TensorPlay

English

简体中文

English

简体中文

Appearance

tensorplay.nn.modules.pooling ​

Classes ​

class AdaptiveAvgPool1d [source] ​

python
AdaptiveAvgPool1d(output_size: Union[int, NoneType, tuple[Optional[int], ...]]) -> None

Bases: _AdaptiveAvgPoolNd

Applies a 1D adaptive average pooling over an input signal composed of several input planes.

The output size is Lout, for any input size. The number of output features is equal to the number of input planes.

Args ​

  • output_size: the target output size Lout.

Shape ​

  • Input: (N,C,Lin) or (C,Lin).
  • Output: (N,C,Lout) or (C,Lout), where Lout=output_size.

Examples ​

python
# target output size of 5
m = nn.AdaptiveAvgPool1d(5)
input = tensorplay.randn(1, 64, 8)
output = m(input)
Methods

__init__(self, output_size: Union[int, NoneType, tuple[Optional[int], ...]]) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AdaptiveAvgPool2d [source] ​

python
AdaptiveAvgPool2d(output_size: Union[int, NoneType, tuple[Optional[int], ...]]) -> None

Bases: _AdaptiveAvgPoolNd

Applies a 2D adaptive average pooling over an input signal composed of several input planes.

The output is of size H x W, for any input size. The number of output features is equal to the number of input planes.

Args ​

  • output_size: the target output size of the image of the form H x W. Can be a tuple (H, W) or a single H for a square image H x H. H and W can be either a int, or None which means the size will be the same as that of the input.

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win).
  • Output: (N,C,S0,S1) or (C,S0,S1), where S=output_size.

Examples ​

python
# target output size of 5x7
m = nn.AdaptiveAvgPool2d((5, 7))
input = tensorplay.randn(1, 64, 8, 9)
output = m(input)
# target output size of 7x7 (square)
m = nn.AdaptiveAvgPool2d(7)
input = tensorplay.randn(1, 64, 10, 9)
output = m(input)
# target output size of 10x7
m = nn.AdaptiveAvgPool2d((None, 7))
input = tensorplay.randn(1, 64, 10, 9)
output = m(input)
Methods

__init__(self, output_size: Union[int, NoneType, tuple[Optional[int], ...]]) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AdaptiveAvgPool3d [source] ​

python
AdaptiveAvgPool3d(output_size: Union[int, NoneType, tuple[Optional[int], ...]]) -> None

Bases: _AdaptiveAvgPoolNd

Applies a 3D adaptive average pooling over an input signal composed of several input planes.

The output is of size D x H x W, for any input size. The number of output features is equal to the number of input planes.

Args ​

  • output_size: the target output size of the form D x H x W. Can be a tuple (D, H, W) or a single number D for a cube D x D x D. D, H and W can be either a int, or None which means the size will be the same as that of the input.

Shape ​

  • Input: (N,C,Din,Hin,Win) or (C,Din,Hin,Win).
  • Output: (N,C,S0,S1,S2) or (C,S0,S1,S2), where S=output_size.

Examples ​

python
# target output size of 5x7x9
m = nn.AdaptiveAvgPool3d((5, 7, 9))
input = tensorplay.randn(1, 64, 8, 9, 10)
output = m(input)
# target output size of 7x7x7 (cube)
m = nn.AdaptiveAvgPool3d(7)
input = tensorplay.randn(1, 64, 10, 9, 8)
output = m(input)
# target output size of 7x9x8
m = nn.AdaptiveAvgPool3d((7, None, None))
input = tensorplay.randn(1, 64, 10, 9, 8)
output = m(input)
Methods

__init__(self, output_size: Union[int, NoneType, tuple[Optional[int], ...]]) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AdaptiveMaxPool1d [source] ​

python
AdaptiveMaxPool1d(output_size: Union[int, NoneType, tuple[Optional[int], ...]], return_indices: bool = False) -> None

Bases: _AdaptiveMaxPoolNd

Applies a 1D adaptive max pooling over an input signal composed of several input planes.

The output size is Lout, for any input size. The number of output features is equal to the number of input planes.

Args ​

  • output_size: the target output size Lout.
  • return_indices: if True, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool1d. Default: False

Shape ​

  • Input: (N,C,Lin) or (C,Lin).
  • Output: (N,C,Lout) or (C,Lout), where Lout=output_size.

Examples ​

python
# target output size of 5
m = nn.AdaptiveMaxPool1d(5)
input = tensorplay.randn(1, 64, 8)
output = m(input)
Methods

__init__(self, output_size: Union[int, NoneType, tuple[Optional[int], ...]], return_indices: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AdaptiveMaxPool2d [source] ​

python
AdaptiveMaxPool2d(output_size: Union[int, NoneType, tuple[Optional[int], ...]], return_indices: bool = False) -> None

Bases: _AdaptiveMaxPoolNd

Applies a 2D adaptive max pooling over an input signal composed of several input planes.

The output is of size Hout×Wout, for any input size. The number of output features is equal to the number of input planes.

Args ​

  • output_size: the target output size of the image of the form Hout×Wout. Can be a tuple (Hout,Wout) or a single Hout for a square image Hout×Hout. Hout and Wout can be either a int, or None which means the size will be the same as that of the input.
  • return_indices: if True, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool2d. Default: False

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win).
  • Output: (N,C,Hout,Wout) or (C,Hout,Wout), where (Hout,Wout)=output_size.

Examples ​

python
# target output size of 5x7
m = nn.AdaptiveMaxPool2d((5, 7))
input = tensorplay.randn(1, 64, 8, 9)
output = m(input)
# target output size of 7x7 (square)
m = nn.AdaptiveMaxPool2d(7)
input = tensorplay.randn(1, 64, 10, 9)
output = m(input)
# target output size of 10x7
m = nn.AdaptiveMaxPool2d((None, 7))
input = tensorplay.randn(1, 64, 10, 9)
output = m(input)
Methods

__init__(self, output_size: Union[int, NoneType, tuple[Optional[int], ...]], return_indices: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AdaptiveMaxPool3d [source] ​

python
AdaptiveMaxPool3d(output_size: Union[int, NoneType, tuple[Optional[int], ...]], return_indices: bool = False) -> None

Bases: _AdaptiveMaxPoolNd

Applies a 3D adaptive max pooling over an input signal composed of several input planes.

The output is of size Dout×Hout×Wout, for any input size. The number of output features is equal to the number of input planes.

Args ​

  • output_size: the target output size of the image of the form Dout×Hout×Wout. Can be a tuple (Dout,Hout,Wout) or a single Dout for a cube Dout×Dout×Dout. Dout, Hout and Wout can be either a int, or None which means the size will be the same as that of the input.
  • return_indices: if True, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool3d. Default: False

Shape ​

  • Input: (N,C,Din,Hin,Win) or (C,Din,Hin,Win).
  • Output: (N,C,Dout,Hout,Wout) or (C,Dout,Hout,Wout), where (Dout,Hout,Wout)=output_size.

Examples ​

python
# target output size of 5x7x9
m = nn.AdaptiveMaxPool3d((5, 7, 9))
input = tensorplay.randn(1, 64, 8, 9, 10)
output = m(input)
# target output size of 7x7x7 (cube)
m = nn.AdaptiveMaxPool3d(7)
input = tensorplay.randn(1, 64, 10, 9, 8)
output = m(input)
# target output size of 7x9x8
m = nn.AdaptiveMaxPool3d((7, None, None))
input = tensorplay.randn(1, 64, 10, 9, 8)
output = m(input)
Methods

__init__(self, output_size: Union[int, NoneType, tuple[Optional[int], ...]], return_indices: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AvgPool1d [source] ​

python
AvgPool1d(kernel_size: Union[int, tuple[int]], stride: Union[int, tuple[int]] = None, padding: Union[int, tuple[int]] = 0, ceil_mode: bool = False, count_include_pad: bool = True) -> None

Bases: _AvgPoolNd

Applies a 1D average pooling over an input signal composed of several input planes.

In the simplest case, the output value of the layer with input size (N,C,L), output (N,C,Lout) and kernel_size k can be precisely described as:

.. math

python
\text{out}(N_i, C_j, l) = \frac{1}{k} \sum_{m=0}^{k-1}
\text{input}(N_i, C_j, \text{stride} \times l + m)

If padding is non-zero, then the input is implicitly zero-padded on both sides for padding number of points.

Note ​

When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding
or the input. Sliding windows that would start in the right padded region are ignored.

INFO

pad should be at most half of effective kernel size.

The parameters kernel_size, stride, padding can each be an int or a one-element tuple.

Args ​

  • kernel_size: the size of the window
  • stride: the stride of the window. Default value is kernel_size
  • padding: implicit zero padding to be added on both sides
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape
  • count_include_pad: when True, will include the zero-padding in the averaging calculation

Shape ​

  • Input: (N,C,Lin) or (C,Lin).

  • Output: (N,C,Lout) or (C,Lout), where

    .. math
python
L_{out} = \left\lfloor \frac{L_{in} +
    2 \times \text{padding} - \text{kernel\_size}}{\text{stride}} + 1\right\rfloor

Per the note above, if ``ceil_mode`` is True and $(L_{out} - 1) \times \text{stride} \geq L_{in}
+ \text{padding}$, we skip the last window as it would start in the right padded region, resulting in
$L_{out}$ being reduced by one.

Examples ​

python
# pool with window of size=3, stride=2
m = nn.AvgPool1d(3, stride=2)
m(tensorplay.tensor([[[1., 2, 3, 4, 5, 6, 7]]]))
tensor([[[2., 4., 6.]]])
Methods

__init__(self, kernel_size: Union[int, tuple[int]], stride: Union[int, tuple[int]] = None, padding: Union[int, tuple[int]] = 0, ceil_mode: bool = False, count_include_pad: bool = True) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AvgPool2d [source] ​

python
AvgPool2d(kernel_size: Union[int, tuple[int, int]], stride: Union[int, tuple[int, int], NoneType] = None, padding: Union[int, tuple[int, int]] = 0, ceil_mode: bool = False, count_include_pad: bool = True, divisor_override: Optional[int] = None) -> None

Bases: _AvgPoolNd

Applies a 2D average pooling over an input signal composed of several input planes.

In the simplest case, the output value of the layer with input size (N,C,H,W), output (N,C,Hout,Wout) and kernel_size (kH,kW) can be precisely described as:

.. math

python
out(N_i, C_j, h, w)  = \frac{1}{kH * kW} \sum_{m=0}^{kH-1} \sum_{n=0}^{kW-1}
input(N_i, C_j, stride[0] \times h + m, stride[1] \times w + n)

If padding is non-zero, then the input is implicitly zero-padded on both sides for padding number of points.

Note ​

When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding
or the input. Sliding windows that would start in the right padded region are ignored.

INFO

pad should be at most half of effective kernel size.

The parameters kernel_size, stride, padding can either be:

  • a single int or a single-element tuple -- in which case the same value is used for the height and width dimension
  • a tuple of two ints -- in which case, the first int is used for the height dimension, and the second int for the width dimension

Args ​

  • kernel_size: the size of the window
  • stride: the stride of the window. Default value is kernel_size
  • padding: implicit zero padding to be added on both sides
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape
  • count_include_pad: when True, will include the zero-padding in the averaging calculation
  • divisor_override: if specified, it will be used as divisor, otherwise size of the pooling region will be used.

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win).

  • Output: (N,C,Hout,Wout) or (C,Hout,Wout), where

    .. math
python
H_{out} = \left\lfloor\frac{H_{in}  + 2 \times \text{padding}[0] -
      \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor

.. math::
    W_{out} = \left\lfloor\frac{W_{in}  + 2 \times \text{padding}[1] -
      \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor

Per the note above, if ``ceil_mode`` is True and $(H_{out} - 1)\times \text{stride}[0]\geq H_{in}
+ \text{padding}[0]$, we skip the last window as it would start in the bottom padded region,
resulting in $H_{out}$ being reduced by one.

The same applies for $W_{out}$.

Examples ​

python
# pool of square window of size=3, stride=2
m = nn.AvgPool2d(3, stride=2)
# pool of non-square window
m = nn.AvgPool2d((3, 2), stride=(2, 1))
input = tensorplay.randn(20, 16, 50, 32)
output = m(input)
Methods

__init__(self, kernel_size: Union[int, tuple[int, int]], stride: Union[int, tuple[int, int], NoneType] = None, padding: Union[int, tuple[int, int]] = 0, ceil_mode: bool = False, count_include_pad: bool = True, divisor_override: Optional[int] = None) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class AvgPool3d [source] ​

python
AvgPool3d(kernel_size: Union[int, tuple[int, int, int]], stride: Union[int, tuple[int, int, int], NoneType] = None, padding: Union[int, tuple[int, int, int]] = 0, ceil_mode: bool = False, count_include_pad: bool = True, divisor_override: Optional[int] = None) -> None

Bases: _AvgPoolNd

Applies a 3D average pooling over an input signal composed of several input planes.

In the simplest case, the output value of the layer with input size (N,C,D,H,W), output (N,C,Dout,Hout,Wout) and kernel_size (kD,kH,kW) can be precisely described as:

.. math

python
\begin{aligned}
    \text{out}(N_i, C_j, d, h, w) ={} & \sum_{k=0}^{kD-1} \sum_{m=0}^{kH-1} \sum_{n=0}^{kW-1} \\
                                      & \frac{\text{input}(N_i, C_j, \text{stride}[0] \times d + k,
                                              \text{stride}[1] \times h + m, \text{stride}[2] \times w + n)}
                                             {kD \times kH \times kW}
\end{aligned}

If padding is non-zero, then the input is implicitly zero-padded on all three sides for padding number of points.

Note ​

When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding
or the input. Sliding windows that would start in the right padded region are ignored.

INFO

pad should be at most half of effective kernel size.

The parameters kernel_size, stride can either be:

  • a single int -- in which case the same value is used for the depth, height and width dimension
  • a tuple of three ints -- in which case, the first int is used for the depth dimension, the second int for the height dimension and the third int for the width dimension

Args ​

  • kernel_size: the size of the window
  • stride: the stride of the window. Default value is kernel_size
  • padding: implicit zero padding to be added on all three sides
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape
  • count_include_pad: when True, will include the zero-padding in the averaging calculation
  • divisor_override: if specified, it will be used as divisor, otherwise kernel_size will be used

Shape ​

  • Input: (N,C,Din,Hin,Win) or (C,Din,Hin,Win).

  • Output: (N,C,Dout,Hout,Wout) or (C,Dout,Hout,Wout), where

    .. math
python
D_{out} = \left\lfloor\frac{D_{in} + 2 \times \text{padding}[0] -
          \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor

.. math::
    H_{out} = \left\lfloor\frac{H_{in} + 2 \times \text{padding}[1] -
          \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor

.. math::
    W_{out} = \left\lfloor\frac{W_{in} + 2 \times \text{padding}[2] -
          \text{kernel\_size}[2]}{\text{stride}[2]} + 1\right\rfloor

Per the note above, if ``ceil_mode`` is True and $(D_{out} - 1)\times \text{stride}[0]\geq D_{in}
+ \text{padding}[0]$, we skip the last window as it would start in the padded region,
resulting in $D_{out}$ being reduced by one.

The same applies for $W_{out}$ and $H_{out}$.

Examples ​

python
# pool of square window of size=3, stride=2
m = nn.AvgPool3d(3, stride=2)
# pool of non-square window
m = nn.AvgPool3d((3, 2, 2), stride=(2, 1, 2))
input = tensorplay.randn(20, 16, 50, 44, 31)
output = m(input)
Methods

__init__(self, kernel_size: Union[int, tuple[int, int, int]], stride: Union[int, tuple[int, int, int], NoneType] = None, padding: Union[int, tuple[int, int, int]] = 0, ceil_mode: bool = False, count_include_pad: bool = True, divisor_override: Optional[int] = None) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class FractionalMaxPool2d [source] ​

python
FractionalMaxPool2d(kernel_size: Union[int, tuple[int, int]], output_size: Union[int, tuple[int, int], NoneType] = None, output_ratio: Union[float, tuple[float, float], NoneType] = None, return_indices: bool = False, _random_samples=None) -> None

Bases: Module

Applies a 2D fractional max pooling over an input signal composed of several input planes.

Fractional MaxPooling is described in detail in the paper Fractional MaxPooling_ by Ben Graham

The max-pooling operation is applied in kH×kW regions by a stochastic step size determined by the target output size. The number of output features is equal to the number of input planes.

.. note:: Exactly one of output_size or output_ratio must be defined.

Args ​

  • kernel_size: the size of the window to take a max over. Can be a single number k (for a square kernel of k x k) or a tuple (kh, kw)
  • output_size: the target output size of the image of the form oH x oW. Can be a tuple (oH, oW) or a single number oH for a square image oH x oH. Note that we must have kH+oH−1<=Hin and kW+oW−1<=Win
  • output_ratio: If one wants to have an output size as a ratio of the input size, this option can be given. This has to be a number or tuple in the range (0, 1). Note that we must have kH+(output_ratio_H∗Hin)−1<=Hin and kW+(output_ratio_W∗Win)−1<=Win
  • return_indices: if True, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool2d. Default: False

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win).
  • Output: (N,C,Hout,Wout) or (C,Hout,Wout), where (Hout,Wout)=output_size or (Hout,Wout)=output_ratio×(Hin,Win).

Examples ​

python
# pool of square window of size=3, and target output size 13x12
m = nn.FractionalMaxPool2d(3, output_size=(13, 12))
# pool of square window and target output size being half of input image size
m = nn.FractionalMaxPool2d(3, output_ratio=(0.5, 0.5))
input = tensorplay.randn(20, 16, 50, 32)
output = m(input)

.. _Fractional MaxPooling:

Methods

__init__(self, kernel_size: Union[int, tuple[int, int]], output_size: Union[int, tuple[int, int], NoneType] = None, output_ratio: Union[float, tuple[float, float], NoneType] = None, return_indices: bool = False, _random_samples=None) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Define the computation performed at every call.

Should be overridden by all subclasses.

INFO

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class FractionalMaxPool3d [source] ​

python
FractionalMaxPool3d(kernel_size: Union[int, tuple[int, int, int]], output_size: Union[int, tuple[int, int, int], NoneType] = None, output_ratio: Union[float, tuple[float, float, float], NoneType] = None, return_indices: bool = False, _random_samples=None) -> None

Bases: Module

Applies a 3D fractional max pooling over an input signal composed of several input planes.

Fractional MaxPooling is described in detail in the paper Fractional MaxPooling_ by Ben Graham

The max-pooling operation is applied in kT×kH×kW regions by a stochastic step size determined by the target output size. The number of output features is equal to the number of input planes.

.. note:: Exactly one of output_size or output_ratio must be defined.

Args ​

  • kernel_size: the size of the window to take a max over. Can be a single number k (for a square kernel of k x k x k) or a tuple (kt x kh x kw), k must greater than 0.
  • output_size: the target output size of the image of the form oT x oH x oW. Can be a tuple (oT, oH, oW) or a single number oH for a square image oH x oH x oH
  • output_ratio: If one wants to have an output size as a ratio of the input size, this option can be given. This has to be a number or tuple in the range (0, 1)
  • return_indices: if True, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool3d. Default: False

Shape ​

  • Input: (N,C,Tin,Hin,Win) or (C,Tin,Hin,Win).
  • Output: (N,C,Tout,Hout,Wout) or (C,Tout,Hout,Wout), where (Tout,Hout,Wout)=output_size or (Tout,Hout,Wout)=output_ratio×(Tin,Hin,Win)

Examples ​

python
# pool of cubic window of size=3, and target output size 13x12x11
m = nn.FractionalMaxPool3d(3, output_size=(13, 12, 11))
# pool of cubic window and target output size being half of input size
m = nn.FractionalMaxPool3d(3, output_ratio=(0.5, 0.5, 0.5))
input = tensorplay.randn(20, 16, 50, 32, 16)
output = m(input)

.. _Fractional MaxPooling:

Methods

__init__(self, kernel_size: Union[int, tuple[int, int, int]], output_size: Union[int, tuple[int, int, int], NoneType] = None, output_ratio: Union[float, tuple[float, float, float], NoneType] = None, return_indices: bool = False, _random_samples=None) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Define the computation performed at every call.

Should be overridden by all subclasses.

INFO

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class LPPool1d [source] ​

python
LPPool1d(norm_type: float, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, ceil_mode: bool = False) -> None

Bases: _LPPoolNd

Applies a 1D power-average pooling over an input signal composed of several input planes.

On each window, the function computed is:

.. math

python
f(X) = \sqrt[p]{\sum_{x \in X} x^{p}}
  • At p = ∞, one gets Max Pooling
  • At p = 1, one gets Sum Pooling (which is proportional to Average Pooling)

.. note:: If the sum to the power of p is zero, the gradient of this function is not defined. This implementation will set the gradient to zero in this case.

Args ​

  • kernel_size: a single int, the size of the window
  • stride: a single int, the stride of the window. Default value is kernel_size
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape

Shape ​

  • Input: (N,C,Lin) or (C,Lin).

  • Output: (N,C,Lout) or (C,Lout), where

    .. math
python
L_{out} = \left\lfloor\frac{L_{in} - \text{kernel\_size}}{\text{stride}} + 1\right\rfloor

Examples ​

python
# power-2 pool of window of length 3, with stride 2.
m = nn.LPPool1d(2, 3, stride=2)
input = tensorplay.randn(20, 16, 50)
output = m(input)
Methods

__init__(self, norm_type: float, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, ceil_mode: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class LPPool2d [source] ​

python
LPPool2d(norm_type: float, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, ceil_mode: bool = False) -> None

Bases: _LPPoolNd

Applies a 2D power-average pooling over an input signal composed of several input planes.

On each window, the function computed is:

.. math

python
f(X) = \sqrt[p]{\sum_{x \in X} x^{p}}
  • At p = ∞, one gets Max Pooling
  • At p = 1, one gets Sum Pooling (which is proportional to average pooling)

The parameters kernel_size, stride can either be:

  • a single int -- in which case the same value is used for the height and width dimension
  • a tuple of two ints -- in which case, the first int is used for the height dimension, and the second int for the width dimension

.. note:: If the sum to the power of p is zero, the gradient of this function is not defined. This implementation will set the gradient to zero in this case.

Args ​

  • kernel_size: the size of the window
  • stride: the stride of the window. Default value is kernel_size
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win).

  • Output: (N,C,Hout,Wout) or (C,Hout,Wout), where

    .. math
python
H_{out} = \left\lfloor\frac{H_{in} - \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor

.. math::
    W_{out} = \left\lfloor\frac{W_{in} - \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor

Examples ​

python
# power-2 pool of square window of size=3, stride=2
m = nn.LPPool2d(2, 3, stride=2)
# pool of non-square window of power 1.2
m = nn.LPPool2d(1.2, (3, 2), stride=(2, 1))
input = tensorplay.randn(20, 16, 50, 32)
output = m(input)
Methods

__init__(self, norm_type: float, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, ceil_mode: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class LPPool3d [source] ​

python
LPPool3d(norm_type: float, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, ceil_mode: bool = False) -> None

Bases: _LPPoolNd

Applies a 3D power-average pooling over an input signal composed of several input planes.

On each window, the function computed is:

.. math

python
f(X) = \sqrt[p]{\sum_{x \in X} x^{p}}
  • At p = ∞, one gets Max Pooling
  • At p = 1, one gets Sum Pooling (which is proportional to average pooling)

The parameters kernel_size, stride can either be:

  • a single int -- in which case the same value is used for the height, width and depth dimension
  • a tuple of three ints -- in which case, the first int is used for the depth dimension, the second int for the height dimension and the third int for the width dimension

.. note:: If the sum to the power of p is zero, the gradient of this function is not defined. This implementation will set the gradient to zero in this case.

Args ​

  • kernel_size: the size of the window
  • stride: the stride of the window. Default value is kernel_size
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape

Shape ​

  • Input: (N,C,Din,Hin,Win) or (C,Din,Hin,Win).

  • Output: (N,C,Dout,Hout,Wout) or (C,Dout,Hout,Wout), where

    .. math
python
D_{out} = \left\lfloor\frac{D_{in} - \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor

.. math::
    H_{out} = \left\lfloor\frac{H_{in} - \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor

.. math::
    W_{out} = \left\lfloor\frac{W_{in} - \text{kernel\_size}[2]}{\text{stride}[2]} + 1\right\rfloor

Examples ​

python
# power-2 pool of square window of size=3, stride=2
m = nn.LPPool3d(2, 3, stride=2)
# pool of non-square window of power 1.2
m = nn.LPPool3d(1.2, (3, 2, 2), stride=(2, 1, 2))
input = tensorplay.randn(20, 16, 50, 44, 31)
output = m(input)
Methods

__init__(self, norm_type: float, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, ceil_mode: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class MaxPool1d [source] ​

python
MaxPool1d(kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, padding: Union[int, tuple[int, ...]] = 0, dilation: Union[int, tuple[int, ...]] = 1, return_indices: bool = False, ceil_mode: bool = False) -> None

Bases: _MaxPoolNd

Applies a 1D max pooling over an input signal composed of several input planes.

In the simplest case, the output value of the layer with input size (N,C,L) and output (N,C,Lout) can be precisely described as:

.. math

python
out(N_i, C_j, k) = \max_{m=0, \ldots, \text{kernel\_size} - 1}
input(N_i, C_j, stride \times k + m)

If padding is non-zero, then the input is implicitly padded with negative infinity on both sides for padding number of points. dilation is the stride between the elements within the sliding window. This link_ has a nice visualization of the pooling parameters.

Note ​

When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding
or the input. Sliding windows that would start in the right padded region are ignored.

Args ​

  • kernel_size: The size of the sliding window, must be > 0.
  • stride: The stride of the sliding window, must be > 0. Default value is kernel_size.
  • padding: Implicit negative infinity padding to be added on both sides, must be >= 0 and <= kernel_size / 2.
  • dilation: The stride between elements within a sliding window, must be > 0.
  • return_indices: If True, will return the argmax along with the max values. Useful for tensorplay.nn.MaxUnpool1d later
  • ceil_mode: If True, will use ceil instead of floor to compute the output shape. This ensures that every element in the input tensor is covered by a sliding window.

Shape ​

  • Input: (N,C,Lin) or (C,Lin).

  • Output: (N,C,Lout) or (C,Lout),

    where ``ceil_mode = False``

.. math
python
L_{out} = \left\lfloor \frac{L_{in} + 2 \times \text{padding} - \text{dilation}
           \times (\text{kernel\_size} - 1) - 1}{\text{stride}}\right\rfloor + 1

  where ``ceil_mode = True``

  .. math::
      L_{out} = \left\lceil \frac{L_{in} + 2 \times \text{padding} - \text{dilation}
            \times (\text{kernel\_size} - 1) - 1 + (stride - 1)}{\text{stride}}\right\rceil + 1

- Ensure that the last pooling starts inside the image, make $L_{out} = L_{out} - 1$
  when $(L_{out} - 1) * \text{stride} >= L_{in} + \text{padding}$.

Examples ​

python
# pool of size=3, stride=2
m = nn.MaxPool1d(3, stride=2)
input = tensorplay.randn(20, 16, 50)
output = m(input)
Methods

__init__(self, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, padding: Union[int, tuple[int, ...]] = 0, dilation: Union[int, tuple[int, ...]] = 1, return_indices: bool = False, ceil_mode: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class MaxPool2d [source] ​

python
MaxPool2d(kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, padding: Union[int, tuple[int, ...]] = 0, dilation: Union[int, tuple[int, ...]] = 1, return_indices: bool = False, ceil_mode: bool = False) -> None

Bases: _MaxPoolNd

Applies a 2D max pooling over an input signal composed of several input planes.

In the simplest case, the output value of the layer with input size (N,C,H,W), output (N,C,Hout,Wout) and kernel_size (kH,kW) can be precisely described as:

.. math

python
\begin{aligned}
    out(N_i, C_j, h, w) ={} & \max_{m=0, \ldots, kH-1} \max_{n=0, \ldots, kW-1} \\
                            & \text{input}(N_i, C_j, \text{stride[0]} \times h + m,
                                           \text{stride[1]} \times w + n)
\end{aligned}

If padding is non-zero, then the input is implicitly padded with negative infinity on both sides for padding number of points. dilation controls the spacing between the kernel points. It is harder to describe, but this link_ has a nice visualization of what dilation does.

Note ​

When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding
or the input. Sliding windows that would start in the right padded region are ignored.

The parameters kernel_size, stride, padding, dilation can either be:

  • a single int -- in which case the same value is used for the height and width dimension
  • a tuple of two ints -- in which case, the first int is used for the height dimension, and the second int for the width dimension

Args ​

  • kernel_size: the size of the window to take a max over
  • stride: the stride of the window. Default value is kernel_size
  • padding: Implicit negative infinity padding to be added on both sides
  • dilation: a parameter that controls the stride of elements in the window
  • return_indices: if True, will return the max indices along with the outputs. Useful for tensorplay.nn.MaxUnpool2d later
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win)

  • Output: (N,C,Hout,Wout) or (C,Hout,Wout), where

    .. math
python
H_{out} = \left\lfloor\frac{H_{in} + 2 * \text{padding[0]} - \text{dilation[0]}
          \times (\text{kernel\_size[0]} - 1) - 1}{\text{stride[0]}} + 1\right\rfloor

.. math::
    W_{out} = \left\lfloor\frac{W_{in} + 2 * \text{padding[1]} - \text{dilation[1]}
          \times (\text{kernel\_size[1]} - 1) - 1}{\text{stride[1]}} + 1\right\rfloor

Examples ​

python
# pool of square window of size=3, stride=2
m = nn.MaxPool2d(3, stride=2)
# pool of non-square window
m = nn.MaxPool2d((3, 2), stride=(2, 1))
input = tensorplay.randn(20, 16, 50, 32)
output = m(input)
Methods

__init__(self, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, padding: Union[int, tuple[int, ...]] = 0, dilation: Union[int, tuple[int, ...]] = 1, return_indices: bool = False, ceil_mode: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class MaxPool3d [source] ​

python
MaxPool3d(kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, padding: Union[int, tuple[int, ...]] = 0, dilation: Union[int, tuple[int, ...]] = 1, return_indices: bool = False, ceil_mode: bool = False) -> None

Bases: _MaxPoolNd

Applies a 3D max pooling over an input signal composed of several input planes.

In the simplest case, the output value of the layer with input size (N,C,D,H,W), output (N,C,Dout,Hout,Wout) and kernel_size (kD,kH,kW) can be precisely described as:

.. math

python
\begin{aligned}
    \text{out}(N_i, C_j, d, h, w) ={} & \max_{k=0, \ldots, kD-1} \max_{m=0, \ldots, kH-1} \max_{n=0, \ldots, kW-1} \\
                                      & \text{input}(N_i, C_j, \text{stride[0]} \times d + k,
                                                     \text{stride[1]} \times h + m, \text{stride[2]} \times w + n)
\end{aligned}

If padding is non-zero, then the input is implicitly padded with negative infinity on both sides for padding number of points. dilation controls the spacing between the kernel points. It is harder to describe, but this link_ has a nice visualization of what dilation does.

Note ​

When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding
or the input. Sliding windows that would start in the right padded region are ignored.

The parameters kernel_size, stride, padding, dilation can either be:

  • a single int -- in which case the same value is used for the depth, height and width dimension
  • a tuple of three ints -- in which case, the first int is used for the depth dimension, the second int for the height dimension and the third int for the width dimension

Args ​

  • kernel_size: the size of the window to take a max over
  • stride: the stride of the window. Default value is kernel_size
  • padding: Implicit negative infinity padding to be added on all three sides
  • dilation: a parameter that controls the stride of elements in the window
  • return_indices: if True, will return the max indices along with the outputs. Useful for tensorplay.nn.MaxUnpool3d later
  • ceil_mode: when True, will use ceil instead of floor to compute the output shape

Shape ​

  • Input: (N,C,Din,Hin,Win) or (C,Din,Hin,Win).

  • Output: (N,C,Dout,Hout,Wout) or (C,Dout,Hout,Wout), where

    .. math
python
D_{out} = \left\lfloor\frac{D_{in} + 2 \times \text{padding}[0] - \text{dilation}[0] \times
      (\text{kernel\_size}[0] - 1) - 1}{\text{stride}[0]} + 1\right\rfloor

.. math::
    H_{out} = \left\lfloor\frac{H_{in} + 2 \times \text{padding}[1] - \text{dilation}[1] \times
      (\text{kernel\_size}[1] - 1) - 1}{\text{stride}[1]} + 1\right\rfloor

.. math::
    W_{out} = \left\lfloor\frac{W_{in} + 2 \times \text{padding}[2] - \text{dilation}[2] \times
      (\text{kernel\_size}[2] - 1) - 1}{\text{stride}[2]} + 1\right\rfloor

Examples ​

python
# pool of square window of size=3, stride=2
m = nn.MaxPool3d(3, stride=2)
# pool of non-square window
m = nn.MaxPool3d((3, 2, 2), stride=(2, 1, 2))
input = tensorplay.randn(20, 16, 50, 44, 31)
output = m(input)
Methods

__init__(self, kernel_size: Union[int, tuple[int, ...]], stride: Union[int, tuple[int, ...], NoneType] = None, padding: Union[int, tuple[int, ...]] = 0, dilation: Union[int, tuple[int, ...]] = 1, return_indices: bool = False, ceil_mode: bool = False) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase) [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class MaxUnpool1d [source] ​

python
MaxUnpool1d(kernel_size: Union[int, tuple[int]], stride: Union[int, tuple[int], NoneType] = None, padding: Union[int, tuple[int]] = 0) -> None

Bases: _MaxUnpoolNd

Computes a partial inverse of MaxPool1d.

MaxPool1d is not fully invertible, since the non-maximal values are lost.

MaxUnpool1d takes in as input the output of MaxPool1d including the indices of the maximal values and computes a partial inverse in which all non-maximal values are set to zero.

Note ​

This operation may behave nondeterministically when the input indices has repeat values.

.. note:: MaxPool1d can map several input sizes to the same output sizes. Hence, the inversion process can get ambiguous. To accommodate this, you can provide the needed output size as an additional argument output_size in the forward call. See the Inputs and Example below.

Args ​

  • kernel_size (int or tuple): Size of the max pooling window.
  • stride (int or tuple): Stride of the max pooling window. It is set to kernel_size by default.
  • padding (int or tuple): Padding that was added to the input

Inputs:

  • input: the input Tensor to invert
  • indices: the indices given out by ~tensorplay.nn.MaxPool1d
  • output_size (optional): the targeted output size

Shape ​

  • Input: (N,C,Hin) or (C,Hin).

  • Output: (N,C,Hout) or (C,Hout), where

    .. math
python
H_{out} = (H_{in} - 1) \times \text{stride}[0] - 2 \times \text{padding}[0] + \text{kernel\_size}[0]

or as given by `output_size` in the call operator

Example ​

python
# xdoctest: +IGNORE_WANT("do other tests modify the global state?")
pool = nn.MaxPool1d(2, stride=2, return_indices=True)
unpool = nn.MaxUnpool1d(2, stride=2)
input = tensorplay.tensor([[[1., 2, 3, 4, 5, 6, 7, 8]]])
output, indices = pool(input)
unpool(output, indices)
tensor([[[ 0.,  2.,  0.,  4.,  0.,  6.,  0., 8.]]])

# Example showcasing the use of output_size
input = tensorplay.tensor([[[1., 2, 3, 4, 5, 6, 7, 8, 9]]])
output, indices = pool(input)
unpool(output, indices, output_size=input.size())
tensor([[[ 0.,  2.,  0.,  4.,  0.,  6.,  0., 8.,  0.]]])

unpool(output, indices)
tensor([[[ 0.,  2.,  0.,  4.,  0.,  6.,  0., 8.]]])
Methods

__init__(self, kernel_size: Union[int, tuple[int]], stride: Union[int, tuple[int], NoneType] = None, padding: Union[int, tuple[int]] = 0) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase, indices: tensorplay._C.TensorBase, output_size: Optional[list[int]] = None) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class MaxUnpool2d [source] ​

python
MaxUnpool2d(kernel_size: Union[int, tuple[int, int]], stride: Union[int, tuple[int, int], NoneType] = None, padding: Union[int, tuple[int, int]] = 0) -> None

Bases: _MaxUnpoolNd

Computes a partial inverse of MaxPool2d.

MaxPool2d is not fully invertible, since the non-maximal values are lost.

MaxUnpool2d takes in as input the output of MaxPool2d including the indices of the maximal values and computes a partial inverse in which all non-maximal values are set to zero.

Note ​

This operation may behave nondeterministically when the input indices has repeat values.

.. note:: MaxPool2d can map several input sizes to the same output sizes. Hence, the inversion process can get ambiguous. To accommodate this, you can provide the needed output size as an additional argument output_size in the forward call. See the Inputs and Example below.

Args ​

  • kernel_size (int or tuple): Size of the max pooling window.
  • stride (int or tuple): Stride of the max pooling window. It is set to kernel_size by default.
  • padding (int or tuple): Padding that was added to the input

Inputs:

  • input: the input Tensor to invert
  • indices: the indices given out by ~tensorplay.nn.MaxPool2d
  • output_size (optional): the targeted output size

Shape ​

  • Input: (N,C,Hin,Win) or (C,Hin,Win).

  • Output: (N,C,Hout,Wout) or (C,Hout,Wout), where

    .. math
python
H_{out} = (H_{in} - 1) \times \text{stride[0]} - 2 \times \text{padding[0]} + \text{kernel\_size[0]}

.. math::
  W_{out} = (W_{in} - 1) \times \text{stride[1]} - 2 \times \text{padding[1]} + \text{kernel\_size[1]}

or as given by `output_size` in the call operator

Example ​

python
pool = nn.MaxPool2d(2, stride=2, return_indices=True)
unpool = nn.MaxUnpool2d(2, stride=2)
input = tensorplay.tensor([[[[ 1.,  2.,  3.,  4.],
                            [ 5.,  6.,  7.,  8.],
                            [ 9., 10., 11., 12.],
                            [13., 14., 15., 16.]]]])
output, indices = pool(input)
unpool(output, indices)
tensor([[[[  0.,   0.,   0.,   0.],
          [  0.,   6.,   0.,   8.],
          [  0.,   0.,   0.,   0.],
          [  0.,  14.,   0.,  16.]]]])
# Now using output_size to resolve an ambiguous size for the inverse
input = tensorplay.tensor([[[[ 1.,  2.,  3.,  4.,  5.],
                            [ 6.,  7.,  8.,  9., 10.],
                            [11., 12., 13., 14., 15.],
                            [16., 17., 18., 19., 20.]]]])
output, indices = pool(input)
# This call will not work without specifying output_size
unpool(output, indices, output_size=input.size())
tensor([[[[ 0.,  0.,  0.,  0.,  0.],
          [ 0.,  7.,  0.,  9.,  0.],
          [ 0.,  0.,  0.,  0.,  0.],
          [ 0., 17.,  0., 19.,  0.]]]])
Methods

__init__(self, kernel_size: Union[int, tuple[int, int]], stride: Union[int, tuple[int, int], NoneType] = None, padding: Union[int, tuple[int, int]] = 0) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase, indices: tensorplay._C.TensorBase, output_size: Optional[list[int]] = None) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

class MaxUnpool3d [source] ​

python
MaxUnpool3d(kernel_size: Union[int, tuple[int, int, int]], stride: Union[int, tuple[int, int, int], NoneType] = None, padding: Union[int, tuple[int, int, int]] = 0) -> None

Bases: _MaxUnpoolNd

Computes a partial inverse of MaxPool3d.

MaxPool3d is not fully invertible, since the non-maximal values are lost. MaxUnpool3d takes in as input the output of MaxPool3d including the indices of the maximal values and computes a partial inverse in which all non-maximal values are set to zero.

Note ​

This operation may behave nondeterministically when the input indices has repeat values.

.. note:: MaxPool3d can map several input sizes to the same output sizes. Hence, the inversion process can get ambiguous. To accommodate this, you can provide the needed output size as an additional argument output_size in the forward call. See the Inputs section below.

Args ​

  • kernel_size (int or tuple): Size of the max pooling window.
  • stride (int or tuple): Stride of the max pooling window. It is set to kernel_size by default.
  • padding (int or tuple): Padding that was added to the input

Inputs:

  • input: the input Tensor to invert
  • indices: the indices given out by ~tensorplay.nn.MaxPool3d
  • output_size (optional): the targeted output size

Shape ​

  • Input: (N,C,Din,Hin,Win) or (C,Din,Hin,Win).

  • Output: (N,C,Dout,Hout,Wout) or (C,Dout,Hout,Wout), where

    .. math
python
D_{out} = (D_{in} - 1) \times \text{stride[0]} - 2 \times \text{padding[0]} + \text{kernel\_size[0]}

.. math::
    H_{out} = (H_{in} - 1) \times \text{stride[1]} - 2 \times \text{padding[1]} + \text{kernel\_size[1]}

.. math::
    W_{out} = (W_{in} - 1) \times \text{stride[2]} - 2 \times \text{padding[2]} + \text{kernel\_size[2]}

or as given by `output_size` in the call operator

Example ​

python
# pool of square window of size=3, stride=2
pool = nn.MaxPool3d(3, stride=2, return_indices=True)
unpool = nn.MaxUnpool3d(3, stride=2)
output, indices = pool(tensorplay.randn(20, 16, 51, 33, 15))
unpooled_output = unpool(output, indices)
unpooled_output.size()
tensorplay.Size([20, 16, 51, 33, 15])
Methods

__init__(self, kernel_size: Union[int, tuple[int, int, int]], stride: Union[int, tuple[int, int, int], NoneType] = None, padding: Union[int, tuple[int, int, int]] = 0) -> None [source] ​

Initialize internal Module state, shared by both nn.Module and ScriptModule.

add_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Add a child module to the current module.

The module can be accessed as an attribute using the given name.

Args ​

  • name (str): name of the child module. The child module can be accessed from this module using the given name
  • module (Module): child module to be added to the module.

apply(self, fn: Callable[[ForwardRef('Module')], NoneType]) -> Self [source] ​

Apply fn recursively to every submodule (as returned by .children()) as well as self.

Typical use includes initializing the parameters of a model (see also nn-init-doc).

Args ​

  • fn (``Module -> None): function to be applied to each submodule

Returns ​

  • Module: self

Example ​

python
@tensorplay.no_grad()
def init_weights(m):
    print(m)
    if type(m) == nn.Linear:
        m.weight.fill_(1.0)
        print(m.weight)
net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
        [1., 1.]], requires_grad=True)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)

bfloat16(self) -> Self [source] ​

Casts all floating point parameters and buffers to bfloat16 datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

buffers(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay._C.TensorBase] [source] ​

Return an iterator over module buffers.

Args ​

  • recurse (bool): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module.

Yields ​

tensorplay.Tensor: module buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for buf in model.buffers():
    print(type(buf), buf.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

children(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over immediate children modules.

Yields ​

  • Module: a child module

compile(self, *args, **kwargs) [source] ​

Compile this Module's forward using tensorplay.compile.

This Module's __call__ method is compiled and all arguments are passed as-is to tensorplay.compile.

See tensorplay.compile for details on the arguments for this function.

cpu(self) -> Self [source] ​

Move all model parameters and buffers to the CPU.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

cuda(self, device: Union[int, tensorplay.Device, NoneType] = None) -> Self [source] ​

Move all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing the optimizer if the module will live on GPU while being optimized.

INFO

This method modifies the module in-place.

Args ​

  • device (int, optional): if specified, all parameters will be copied to that device

Returns ​

  • Module: self

double(self) -> Self [source] ​

Casts all floating point parameters and buffers to double datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

eval(self) -> Self [source] ​

Set the module in evaluation mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e. whether they are affected, e.g. Dropout, BatchNorm, etc.

This is equivalent with self.train(False) <tensorplay.nn.Module.train>.

See locally-disable-grad-doc for a comparison between .eval() and several similar mechanisms that may be confused with it.

Returns ​

  • Module: self

extra_repr(self) -> str [source] ​

Return the extra representation of the module.

To print customized extra information, you should re-implement this method in your own modules. Both single-line and multi-line strings are acceptable.

float(self) -> Self [source] ​

Casts all floating point parameters and buffers to float datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

forward(self, input: tensorplay._C.TensorBase, indices: tensorplay._C.TensorBase, output_size: Optional[list[int]] = None) -> tensorplay._C.TensorBase [source] ​

Runs the forward pass.

get_buffer(self, target: str) -> 'Tensor' [source] ​

Return the buffer given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the buffer to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.Tensor: The buffer referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not a buffer

get_extra_state(self) -> Any [source] ​

Return any extra state to include in the module's state_dict.

Implement this and a corresponding set_extra_state for your module if you need to store extra state. This function is called when building the module's state_dict().

Note that extra state should be picklable to ensure working serialization of the state_dict. We only provide backwards compatibility guarantees for serializing Tensors; other objects may break backwards compatibility if their serialized pickled form changes.

Returns ​

  • object: Any extra state to store in the module's state_dict

get_parameter(self, target: str) -> 'Parameter' [source] ​

Return the parameter given by target if it exists, otherwise throw an error.

See the docstring for get_submodule for a more detailed explanation of this method's functionality as well as how to correctly specify target.

Args ​

  • target: The fully-qualified string name of the Parameter to look for. (See get_submodule for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Parameter: The Parameter referenced by ``target``

Raises ​

  • AttributeError: If the target string references an invalid path or resolves to something that is not an nn.Parameter

get_submodule(self, target: str) -> 'Module' [source] ​

Return the submodule given by target if it exists, otherwise throw an error.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2)) ) (linear): Linear(in_features=100, out_features=200, bias=True) ) )

(The diagram shows an nn.Module A. A which has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To check whether or not we have the linear submodule, we would call get_submodule("net_b.linear"). To check whether we have the conv submodule, we would call get_submodule("net_b.net_c.conv").

The runtime of get_submodule is bounded by the degree of module nesting in target. A query against named_modules achieves the same result, but it is O(N) in the number of transitive modules. So, for a simple check to see if some submodule exists, get_submodule should always be used.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)

Returns ​

tensorplay.nn.Module: The submodule referenced by ``target``

Raises ​

  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

half(self) -> Self [source] ​

Casts all floating point parameters and buffers to half datatype.

INFO

This method modifies the module in-place.

Returns ​

  • Module: self

load_state_dict(self, state_dict: collections.abc.Mapping[str, typing.Any], strict: bool = True, assign: bool = False) [source] ​

Copy parameters and buffers from state_dict into this module and its descendants.

If strict is True, then the keys of state_dict must exactly match the keys returned by this module's ~tensorplay.nn.Module.state_dict function.

DANGER

If assign is True the optimizer must be created after the call to load_state_dict unless ~tensorplay.__future__.get_swap_module_params_on_conversion is True.

Args ​

  • state_dict (dict): a dict containing parameters and persistent buffers.
  • strict (bool, optional): whether to strictly enforce that the keys in state_dict match the keys returned by this module's ~tensorplay.nn.Module.state_dict function. Default: True
  • assign (bool, optional): When set to False, the properties of the tensors in the current module are preserved whereas setting it to True preserves properties of the Tensors in the state dict. The only exception is the requires_grad field of ~tensorplay.nn.Parameter for which the value from the module is preserved. Default: False

Returns ​

``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:
    * ``missing_keys`` is a list of str containing any keys that are expected
        by this module but missing from the provided ``state_dict``.
    * ``unexpected_keys`` is a list of str containing the keys that are not
        expected by this module but present in the provided ``state_dict``.

Note ​

If a parameter or buffer is registered as ``None`` and its corresponding key
exists in `state_dict`, `load_state_dict` will raise a
``RuntimeError``.

modules(self) -> collections.abc.Iterator['Module'] [source] ​

Return an iterator over all modules in the network.

Yields ​

  • Module: a module in the network

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.modules()):
    print(idx, '->', m)

0 -> Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
1 -> Linear(in_features=2, out_features=2, bias=True)

named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay._C.TensorBase]] [source] ​

Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Args ​

  • prefix (str): prefix to prepend to all buffer names.
  • recurse (bool, optional): if True, then yields buffers of this module and all submodules. Otherwise, yields only buffers that are direct members of this module. Defaults to True.
  • remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.

Yields ​

(str, tensorplay.Tensor): Tuple containing the name and buffer

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, buf in self.named_buffers():
    if name in ['running_var']:
        print(buf.size())

named_children(self) -> collections.abc.Iterator[tuple[str, 'Module']] [source] ​

Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields ​

(str, Module): Tuple containing a name and child module

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, module in model.named_children():
    if name in ['conv4', 'conv5']:
        print(module)

named_modules(self, memo: Optional[set['Module']] = None, prefix: str = '', remove_duplicate: bool = True) [source] ​

Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Args ​

  • memo: a memo to store the set of modules already added to the result
  • prefix: a prefix that will be added to the name of the module
  • remove_duplicate: whether to remove the duplicated module instances in the result or not

Yields ​

(str, Module): Tuple of name and module

Note ​

Duplicate modules are returned only once. In the following
example, ``l`` will be returned only once.

Example ​

python
l = nn.Linear(2, 2)
net = nn.Sequential(l, l)
for idx, m in enumerate(net.named_modules()):
    print(idx, '->', m)

0 -> ('', Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
))
1 -> ('0', Linear(in_features=2, out_features=2, bias=True))

named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> collections.abc.Iterator[tuple[str, tensorplay.nn.parameter.Parameter]] [source] ​

Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.

Args ​

  • prefix (str): prefix to prepend to all parameter names.
  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.
  • remove_duplicate (bool, optional): whether to remove the duplicated parameters in the result. Defaults to True.

Yields ​

(str, Parameter): Tuple containing the name and parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for name, param in self.named_parameters():
    if name in ['bias']:
        print(param.size())

parameters(self, recurse: bool = True) -> collections.abc.Iterator[tensorplay.nn.parameter.Parameter] [source] ​

Return an iterator over module parameters.

This is typically passed to an optimizer.

Args ​

  • recurse (bool): if True, then yields parameters of this module and all submodules. Otherwise, yields only parameters that are direct members of this module.

Yields ​

  • Parameter: module parameter

Example ​

python
# xdoctest: +SKIP("undefined vars")
for param in model.parameters():
    print(type(param), param.size())
<class 'tensorplay.Tensor'> (20L,)
<class 'tensorplay.Tensor'> (20L, 1L, 5L, 5L)

register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]]) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

This function is deprecated in favor of ~tensorplay.nn.Module.register_full_backward_hook and the behavior of this function will change in future versions.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_buffer(self, name: str, tensor: Optional[tensorplay._C.TensorBase], persistent: bool = True) -> None [source] ​

Add a buffer to the module.

This is typically used to register a buffer that should not be considered a model parameter. For example, BatchNorm's running_mean is not a parameter, but is part of the module's state. Buffers, by default, are persistent and will be saved alongside parameters. This behavior can be changed by setting persistent to False. The only difference between a persistent buffer and a non-persistent buffer is that the latter will not be a part of this module's state_dict.

Buffers can be accessed as attributes using given names.

Args ​

  • name (str): name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor or None): buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored. If None, the buffer is not included in the module's state_dict.
  • persistent (bool): whether the buffer is part of this module's state_dict.

Example ​

python
# xdoctest: +SKIP("undefined vars")
self.register_buffer('running_mean', tensorplay.zeros(num_features))

register_forward_hook(self, hook: Union[Callable[[~T, tuple[Any, ...], Any], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any], Any], Optional[Any]]], *, prepend: bool = False, with_kwargs: bool = False, always_call: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward hook on the module.

The hook will be called every time after forward has computed an output.

If with_kwargs is False or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the output. It can modify the input inplace but it will not have effect on forward since this is called after forward is called. The hook should have the following signature

python
hook(module, args, output) -> None or modified output

If with_kwargs is True, the forward hook will be passed the kwargs given to the forward function and be expected to return the output possibly modified. The hook should have the following signature

python
hook(module, args, kwargs, output) -> None or modified output

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If True, the provided hook will be fired before all existing forward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward hooks on this tensorplay.nn.Module. Note that global forward hooks registered with register_module_forward_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If True, the hook will be passed the kwargs given to the forward function.
  • Default: False
  • always_call (bool): If True the hook will be run regardless of whether an exception is raised while calling the Module.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_forward_pre_hook(self, hook: Union[Callable[[~T, tuple[Any, ...]], Optional[Any]], Callable[[~T, tuple[Any, ...], dict[str, Any]], Optional[tuple[Any, dict[str, Any]]]]], *, prepend: bool = False, with_kwargs: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a forward pre-hook on the module.

The hook will be called every time before forward is invoked.

If with_kwargs is false or not specified, the input contains only the positional arguments given to the module. Keyword arguments won't be passed to the hooks and only to the forward. The hook can modify the input. User can either return a tuple or a single modified value in the hook. We will wrap the value into a tuple if a single value is returned (unless that value is already a tuple). The hook should have the following signature

python
hook(module, args) -> None or modified input

If with_kwargs is true, the forward pre-hook will be passed the kwargs given to the forward function. And if the hook modifies the input, both the args and kwargs should be returned. The hook should have the following signature

python
hook(module, args, kwargs) -> None or a tuple of modified input and kwargs

Args ​

  • hook (Callable): The user defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing forward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing forward_pre hooks on this tensorplay.nn.Module. Note that global forward_pre hooks registered with register_module_forward_pre_hook will fire before all hooks registered by this method.
  • Default: False
  • with_kwargs (bool): If true, the hook will be passed the kwargs given to the forward function.
  • Default: False

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase], Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward hook on the module.

The hook will be called every time the gradients with respect to a module are computed, and its firing rules are as follows:

1. Ordinarily, the hook fires when the gradients are computed with respect to the module inputs.
2. If none of the module inputs require gradients, the hook will fire when the gradients are computed
   with respect to module outputs.
3. If none of the module outputs require gradients, then the hooks will not fire.

The hook should have the following signature

python
hook(module, grad_input, grad_output) -> tuple(Tensor) or None

The grad_input and grad_output are tuples that contain the gradients with respect to the inputs and outputs respectively. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the input that will be used in place of grad_input in subsequent computations. grad_input will only correspond to the inputs given as positional arguments and all kwarg arguments are ignored. Entries in grad_input and grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs or outputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward hooks on this tensorplay.nn.Module. Note that global backward hooks registered with register_module_full_backward_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], Union[NoneType, tuple[tensorplay._C.TensorBase, ...], tensorplay._C.TensorBase]], prepend: bool = False) -> tensorplay.utils.hooks.RemovableHandle [source] ​

Register a backward pre-hook on the module.

The hook will be called every time the gradients for the module are computed. The hook should have the following signature

python
hook(module, grad_output) -> tuple[Tensor] or None

The grad_output is a tuple. The hook should not modify its arguments, but it can optionally return a new gradient with respect to the output that will be used in place of grad_output in subsequent computations. Entries in grad_output will be None for all non-Tensor arguments.

For technical reasons, when this hook is applied to a Module, its forward function will receive a view of each Tensor passed to the Module. Similarly the caller will receive a view of each Tensor returned by the Module's forward function.

.. warning

python
Modifying inputs inplace is not allowed when using backward hooks and
will raise an error.

Args ​

  • hook (Callable): The user-defined hook to be registered.
  • prepend (bool): If true, the provided hook will be fired before all existing backward_pre hooks on this tensorplay.nn.Module. Otherwise, the provided hook will be fired after all existing backward_pre hooks on this tensorplay.nn.Module. Note that global backward_pre hooks registered with register_module_full_backward_pre_hook will fire before all hooks registered by this method.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_post_hook(self, hook) [source] ​

Register a post-hook to be run after module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, incompatible_keys) -> None

The module argument is the current module that this hook is registered on, and the incompatible_keys argument is a NamedTuple consisting of attributes missing_keys and unexpected_keys. missing_keys is a list of str containing the missing keys and unexpected_keys is a list of str containing the unexpected keys.

The given incompatible_keys can be modified inplace if needed.

Note that the checks performed when calling load_state_dict with strict=True are affected by modifications the hook makes to missing_keys or unexpected_keys, as expected. Additions to either set of keys will result in an error being thrown when strict=True, and clearing out both missing and unexpected keys will avoid an error.

Returns ​

`tensorplay.utils.hooks.RemovableHandle`:
    a handle that can be used to remove the added hook by calling
    ``handle.remove()``

register_load_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook to be run before module's ~nn.Module.load_state_dict is called.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) -> None  # noqa: B950

Arguments ​

  • hook (Callable): Callable hook that will be invoked before loading the state dict.

register_module(self, name: str, module: Optional[ForwardRef('Module')]) -> None [source] ​

Alias for add_module.

register_parameter(self, name: str, param: Optional[tensorplay.nn.parameter.Parameter]) -> None [source] ​

Add a parameter to the module.

The parameter can be accessed as an attribute using given name.

Args ​

  • name (str): name of the parameter. The parameter can be accessed from this module using the given name
  • param (Parameter or None): parameter to be added to the module. If None, then operations that run on parameters, such as cuda, are ignored. If None, the parameter is not included in the module's state_dict.

register_state_dict_post_hook(self, hook) [source] ​

Register a post-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, state_dict, prefix, local_metadata) -> None

The registered hooks can modify the state_dict inplace.

register_state_dict_pre_hook(self, hook) [source] ​

Register a pre-hook for the ~tensorplay.nn.Module.state_dict method.

It should have the following signature

python
hook(module, prefix, keep_vars) -> None

The registered hooks can be used to perform pre-processing before the state_dict call is made.

requires_grad_(self, requires_grad: bool = True) -> Self [source] ​

Change if autograd should record operations on parameters in this module.

This method sets the parameters' requires_grad attributes in-place.

This method is helpful for freezing part of the module for finetuning or training parts of a model individually (e.g., GAN training).

See locally-disable-grad-doc for a comparison between .requires_grad_() and several similar mechanisms that may be confused with it.

Args ​

  • requires_grad (bool): whether autograd should record operations on parameters in this module. Default: True.

Returns ​

  • Module: self

set_extra_state(self, state: Any) -> None [source] ​

Set extra state contained in the loaded state_dict.

This function is called from load_state_dict to handle any extra state found within the state_dict. Implement this function and a corresponding get_extra_state for your module if you need to store extra state within its state_dict.

Args ​

  • state (dict): Extra state from the state_dict

set_submodule(self, target: str, module: 'Module', strict: bool = False) -> None [source] ​

Set the submodule given by target if it exists, otherwise throw an error.

INFO

If strict is set to False (default), the method will replace an existing submodule or create a new submodule if the parent module exists. If strict is set to True, the method will only attempt to replace an existing submodule and throw an error if the submodule does not exist.

For example, let's say you have an nn.Module A that looks like this:

.. code-block:: text

  • A ( (net_b): Module( (net_c): Module( (conv): Conv2d(3, 3, 3) ) (linear): Linear(3, 3) ) )

(The diagram shows an nn.Module A. A has a nested submodule net_b, which itself has two submodules net_c and linear. net_c then has a submodule conv.)

To override the Conv2d with a new submodule Linear, you could call set_submodule("net_b.net_c.conv", nn.Linear(1, 1)) where strict could be True or False

To add a new submodule Conv2d to the existing net_b module, you would call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1)).

In the above if you set strict=True and call set_submodule("net_b.conv", nn.Conv2d(1, 1, 1), strict=True), an AttributeError will be raised because net_b does not have a submodule named conv.

Args ​

  • target: The fully-qualified string name of the submodule to look for. (See above example for how to specify a fully-qualified string.)
  • module: The module to set the submodule to.
  • strict: If False, the method will replace an existing submodule or create a new submodule if the parent module exists. If True, the method will only attempt to replace an existing submodule and throw an error if the submodule doesn't already exist.

Raises ​

  • ValueError: If the target string is empty or if module is not an instance of nn.Module.
  • AttributeError: If at any point along the path resulting from the target string the (sub)path resolves to a non-existent attribute name or an object that is not an instance of nn.Module.

share_memory(self) -> Self [source] ​

See tensorplay.Tensor.share_memory_.

state_dict(self, *args, destination=None, prefix='', keep_vars=False) [source] ​

Return a dictionary containing references to the whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names. Parameters and buffers set to None are not included.

INFO

The returned object is a shallow copy. It contains references to the module's parameters and buffers.

DANGER

Currently state_dict() also accepts positional arguments for destination, prefix and keep_vars in order. However, this is being deprecated and keyword arguments will be enforced in future releases.

DANGER

Please avoid the use of argument destination as it is not designed for end-users.

Args ​

  • destination (dict, optional): If provided, the state of module will be updated into the dict and the same object is returned. Otherwise, an OrderedDict will be created and returned.
  • Default: None.
  • prefix (str, optional): a prefix added to parameter and buffer names to compose the keys in state_dict. Default: ''.
  • keep_vars (bool, optional): by default the ~tensorplay.Tensor s returned in the state dict are detached from autograd. If it's set to True, detaching will not be performed.
  • Default: False.

Returns ​

  • dict: a dictionary containing a whole state of the module

Example ​

python
# xdoctest: +SKIP("undefined vars")
module.state_dict().keys()
['bias', 'weight']

to(self, *args, **kwargs) [source] ​

Move and/or cast the parameters and buffers.

This can be called as

.. function:: to(device=None, dtype=None, non_blocking=False) :noindex:

.. function:: to(dtype, non_blocking=False) :noindex:

.. function:: to(tensor, non_blocking=False) :noindex:

.. function:: to(memory_format=tensorplay.channels_last) :noindex:

Its signature is similar to tensorplay.Tensor.to, but only accepts floating point or complex dtype\ s. In addition, this method will only cast the floating point or complex parameters and buffers to dtype (if given). The integral parameters and buffers will be moved device, if that is given, but with dtypes unchanged. When non_blocking is set, it tries to convert/move asynchronously with respect to the host if possible, e.g., moving CPU Tensors with pinned memory to CUDA devices.

See below for examples.

INFO

This method modifies the module in-place.

Args ​

  • device (tensorplay.device): the desired device of the parameters and buffers in this module
  • dtype (tensorplay.dtype): the desired floating point or complex dtype of the parameters and buffers in this module
  • tensor (tensorplay.Tensor): Tensor whose dtype and device are the desired dtype and device for all parameters and buffers in this module
  • memory_format (tensorplay.memory_format): the desired memory format for 4D parameters and buffers in this module (keyword only argument)

Returns ​

  • Module: self

Examples ​

python
# xdoctest: +IGNORE_WANT("non-deterministic")
linear = nn.Linear(2, 2)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
linear.to(tensorplay.double)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=tensorplay.float64)
# xdoctest: +REQUIRES(env:TENSORPLAY_DOCTEST_CUDA1)
gpu1 = tensorplay.device("cuda:1")
linear.to(gpu1, dtype=tensorplay.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16, device='cuda:1')
cpu = tensorplay.device("cpu")
linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=tensorplay.float16)

linear = nn.Linear(2, 2, bias=None).to(tensorplay.cdouble)
linear.weight
Parameter containing:
tensor([[ 0.3741+0.j,  0.2382+0.j],
        [ 0.5593+0.j, -0.4443+0.j]], dtype=tensorplay.complex128)
linear(tensorplay.ones(3, 2, dtype=tensorplay.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j],
        [0.6122+0.j, 0.1150+0.j]], dtype=tensorplay.complex128)

to_empty(self, *, device: Union[str, tensorplay.Device, int, NoneType], recurse: bool = True) -> Self [source] ​

Move the parameters and buffers to the specified device without copying storage.

Args ​

  • device (tensorplay.device): The desired device of the parameters and buffers in this module.
  • recurse (bool): Whether parameters and buffers of submodules should be recursively moved to the specified device.

Returns ​

  • Module: self

train(self, mode: bool = True) -> Self [source] ​

Set the module in training mode.

This has an effect only on certain modules. See the documentation of particular modules for details of their behaviors in training/evaluation mode, i.e., whether they are affected, e.g. Dropout, BatchNorm, etc.

Args ​

  • mode (bool): whether to set training mode (True) or evaluation mode (False). Default: True.

Returns ​

  • Module: self

type(self, dst_type: Union[tensorplay.DType, str]) -> Self [source] ​

Casts all parameters and buffers to dst_type.

INFO

This method modifies the module in-place.

Args ​

  • dst_type (type or string): the desired type

Returns ​

  • Module: self

zero_grad(self, set_to_none: bool = True) -> None [source] ​

Reset gradients of all model parameters.

See similar function under tensorplay.optim.Optimizer for more context.

Args ​

  • set_to_none (bool): instead of setting to zero, set the grads to None. See tensorplay.optim.Optimizer.zero_grad for details.

Released under the Apache 2.0 License.

Copyright © 2025 zlx. All rights reserved.

📚DeepWiki