torch.nn.modules.rnn.RNNBase Class Reference

Inheritance diagram for torch.nn.modules.rnn.RNNBase:

Public Member Functions
def	__init__ (self, mode, input_size, hidden_size, num_layers=1, bias=True, batch_first=False, dropout=0., bidirectional=False)
def	flatten_parameters (self)
def	reset_parameters (self)
def	get_flat_weights (self)
def	check_input (self, input, batch_sizes)
def	get_expected_hidden_size (self, input, batch_sizes)
def	check_hidden_size (self, hx, expected_hidden_size, msg='Expected hidden size{}, got)
def	check_forward_args (self, input, hidden, batch_sizes)
def	permute_hidden (self, hx, permutation)
def	forward (self, input, hx=None)
def	extra_repr (self)
def	__setstate__ (self, d)
def	all_weights (self)

Public Attributes
	mode
	input_size
	hidden_size
	num_layers
	bias
	batch_first
	dropout
	bidirectional

Detailed Description

Definition at line 28 of file rnn.py.

Member Function Documentation

def torch.nn.modules.rnn.RNNBase.flatten_parameters

(

self

)

Resets parameter data pointer so that they can use faster code paths.

Right now, this works only if the module is on the GPU and cuDNN is enabled.
Otherwise, it's a no-op.

Definition at line 94 of file rnn.py.

---

The documentation for this class was generated from the following file:

• torch/nn/modules/rnn.py