tensorplay.distributed.nn.functional.all_to_all

tensorplay.distributed.nn.functional.all_to_all(output_tensor_list, input_tensor_list, group=None)[source]