# tensorplay.library

Source: https://www.tensorplay.cn/docs/library.html

# tensorplay.library

## Choosing the kind of custom op

[tensorplay.library.custom_op](generated/tensorplay.library.custom_op.html#tensorplay.library.custom_op)

Define a user operator; use as a decorator (torch parity).

[tensorplay.library.triton_op](generated/tensorplay.library.triton_op.html#tensorplay.library.triton_op)

Define a Triton-backed operator (torch.library.triton_op).

[tensorplay.library.wrap_triton](generated/tensorplay.library.wrap_triton.html#tensorplay.library.wrap_triton)

Mark a Triton kernel as launchable from within a triton_op.

## Extending custom ops (created from Python or C++)

[tensorplay.library.register_kernel](generated/tensorplay.library.register_kernel.html#tensorplay.library.register_kernel)

Top-level kernel registration (torch.library.register_kernel).

[tensorplay.library.register_autograd](generated/tensorplay.library.register_autograd.html#tensorplay.library.register_autograd)

Top-level autograd registration (torch.library.register_autograd).

[tensorplay.library.register_fake](generated/tensorplay.library.register_fake.html#tensorplay.library.register_fake)

Top-level fake-kernel registration (torch.library.register_fake).

## Low-level APIs

[tensorplay.library.Library](generated/tensorplay.library.Library.html#tensorplay.library.Library)

Scoped registration bundle (torch.library.Library subset).