torchrecurrent.WMCLSTM#
- class torchrecurrent.WMCLSTM(input_size, hidden_size, num_layers=1, dropout=0.0, batch_first=False, **kwargs)[source]#
Multi-layer LSTM with working-memory connections (WMCLSTM).
[arXiv]
Each layer consists of a
WMCLSTMCell
, an LSTM variant where the input, forget, and output gates receive additional connections from the cell state. The updates are:\[\begin{split}\begin{aligned} i(t) &= \sigma\bigl(W_{ih}^i x(t) + b_{ih}^i + W_{hh}^i h(t-1) + b_{hh}^i + W_{mh}^i c(t-1) + b_{mh}^i \bigr), \\ f(t) &= \sigma\bigl(W_{ih}^f x(t) + b_{ih}^f + W_{hh}^f h(t-1) + b_{hh}^f + W_{mh}^f c(t-1) + b_{mh}^f \bigr), \\ c(t) &= f(t) \circ c(t-1) + i(t) \circ \sigma_c(W_{ih}^c x(t) + b_{ih}^c), \\ o(t) &= \sigma\bigl(W_{ih}^o x(t) + b_{ih}^o + W_{hh}^o h(t-1) + b_{hh}^o + W_{mh}^o c(t) + b_{mh}^o \bigr), \\ h(t) &= o(t) \circ \sigma_h(c(t)), \end{aligned}\end{split}\]where \(\sigma\) is the sigmoid, \(\sigma_c\) / \(\sigma_h\) are cell/output activations (usually
torch.tanh()
), and \(\circ\) is the elementwise product.- Parameters:
input_size – Number of expected features in the input x.
hidden_size – Number of features in the hidden and cell states.
num_layers – Number of stacked recurrent layers. Default: 1
dropout – If non-zero, adds dropout after each layer (except last). Default: 0
batch_first – If
True
, inputs and outputs are (batch, seq, feature) instead of (seq, batch, feature). Default: Falsebias – If
False
, disables input biases b_{ih}. Default: Truerecurrent_bias – If
False
, disables recurrent biases b_{hh}. Default: Truememory_bias – If
False
, disables memory biases b_{mh}. Default: Truekernel_init – Initializer for W_{ih}. Default:
torch.nn.init.xavier_uniform_()
recurrent_kernel_init – Initializer for W_{hh}. Default:
torch.nn.init.xavier_uniform_()
memory_kernel_init – Initializer for W_{mh}. Default:
torch.nn.init.xavier_uniform_()
bias_init – Initializer for b_{ih}. Default:
torch.nn.init.zeros_()
recurrent_bias_init – Initializer for b_{hh}. Default:
torch.nn.init.zeros_()
memory_bias_init – Initializer for b_{mh}. Default:
torch.nn.init.zeros_()
device – Desired device of parameters.
dtype – Desired floating point type of parameters.
- Inputs: input, (h_0, c_0)
input: tensor of shape (L, H_in) for unbatched input, (L, N, H_in) when
batch_first=False
, or (N, L, H_in) whenbatch_first=True
.h_0: tensor of shape (num_layers, H_out) (unbatched) or (num_layers, N, H_out) containing initial hidden state. Defaults to zeros if not provided.
c_0: tensor of same shape as h_0, containing initial cell state. Defaults to zeros if not provided.
Where:
\[\begin{split}\begin{aligned} N &= \text{batch size} \\ L &= \text{sequence length} \\ H_{in} &= \text{input size} \\ H_{out} &= \text{hidden size} \end{aligned}\end{split}\]- Outputs: output, (h_n, c_n)
output: tensor of shape (L, H_out) for unbatched input, (L, N, H_out) when
batch_first=False
, or (N, L, H_out) whenbatch_first=True
containing hidden states from the last layer at each timestep.h_n: final hidden state for each layer, shape (num_layers, H_out) (unbatched) or (num_layers, N, H_out).
c_n: final cell state for each layer, same shape as h_n.
- cells.{k}.weight_ih
input–hidden weights of the \(k\)-th layer, shape (4*hidden_size, input_size) for k=0, otherwise (4*hidden_size, hidden_size).
- cells.{k}.weight_hh
hidden–hidden weights of the \(k\)-th layer, shape (4*hidden_size, hidden_size) (i, f, o gates).
- cells.{k}.weight_mh
memory–hidden weights of the \(k\)-th layer, shape (3*hidden_size, hidden_size) (i, f depend on c(t-1), o on c(t)).
- cells.{k}.bias_ih
input biases of the \(k\)-th layer, shape (4*hidden_size,) if
bias=True
.
- cells.{k}.bias_hh
hidden biases of the \(k\)-th layer, shape (4*hidden_size,) if
recurrent_bias=True
.
- cells.{k}.bias_mh
memory biases of the \(k\)-th layer, shape (3*hidden_size,) if
memory_bias=True
.
See also
Examples:
>>> rnn = WMCLSTM(8, 16, num_layers=2) >>> x = torch.randn(5, 3, 8) # (seq_len, batch, input_size) >>> h0 = torch.zeros(2, 3, 16) >>> c0 = torch.zeros(2, 3, 16) >>> out, (hn, cn) = rnn(x, (h0, c0))
- __init__(input_size, hidden_size, num_layers=1, dropout=0.0, batch_first=False, **kwargs)[source]#
Initialize internal Module state, shared by both nn.Module and ScriptModule.
Methods
__init__
(input_size, hidden_size[, ...])Initialize internal Module state, shared by both nn.Module and ScriptModule.
add_module
(name, module)Add a child module to the current module.
apply
(fn)Apply
fn
recursively to every submodule (as returned by.children()
) as well as self.bfloat16
()Casts all floating point parameters and buffers to
bfloat16
datatype.buffers
([recurse])Return an iterator over module buffers.
children
()Return an iterator over immediate children modules.
compile
(*args, **kwargs)Compile this Module's forward using
torch.compile()
.cpu
()Move all model parameters and buffers to the CPU.
cuda
([device])Move all model parameters and buffers to the GPU.
double
()Casts all floating point parameters and buffers to
double
datatype.eval
()Set the module in evaluation mode.
extra_repr
()Return the extra representation of the module.
float
()Casts all floating point parameters and buffers to
float
datatype.forward
(inp[, state])Define the computation performed at every call.
get_buffer
(target)Return the buffer given by
target
if it exists, otherwise throw an error.get_extra_state
()Return any extra state to include in the module's state_dict.
get_parameter
(target)Return the parameter given by
target
if it exists, otherwise throw an error.get_submodule
(target)Return the submodule given by
target
if it exists, otherwise throw an error.half
()Casts all floating point parameters and buffers to
half
datatype.initialize_cells
(cell_class, **kwargs)Helper method to initialize cells for the derived recurrent layer class.
ipu
([device])Move all model parameters and buffers to the IPU.
load_state_dict
(state_dict[, strict, assign])Copy parameters and buffers from
state_dict
into this module and its descendants.modules
()Return an iterator over all modules in the network.
mtia
([device])Move all model parameters and buffers to the MTIA.
named_buffers
([prefix, recurse, ...])Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children
()Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules
([memo, prefix, remove_duplicate])Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters
([prefix, recurse, ...])Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters
([recurse])Return an iterator over module parameters.
register_backward_hook
(hook)Register a backward hook on the module.
register_buffer
(name, tensor[, persistent])Add a buffer to the module.
register_forward_hook
(hook, *[, prepend, ...])Register a forward hook on the module.
register_forward_pre_hook
(hook, *[, ...])Register a forward pre-hook on the module.
register_full_backward_hook
(hook[, prepend])Register a backward hook on the module.
register_full_backward_pre_hook
(hook[, prepend])Register a backward pre-hook on the module.
register_load_state_dict_post_hook
(hook)Register a post-hook to be run after module's
load_state_dict()
is called.register_load_state_dict_pre_hook
(hook)Register a pre-hook to be run before module's
load_state_dict()
is called.register_module
(name, module)Alias for
add_module()
.register_parameter
(name, param)Add a parameter to the module.
register_state_dict_post_hook
(hook)Register a post-hook for the
state_dict()
method.register_state_dict_pre_hook
(hook)Register a pre-hook for the
state_dict()
method.requires_grad_
([requires_grad])Change if autograd should record operations on parameters in this module.
set_extra_state
(state)Set extra state contained in the loaded state_dict.
set_submodule
(target, module[, strict])Set the submodule given by
target
if it exists, otherwise throw an error.share_memory
()See
torch.Tensor.share_memory_()
.state_dict
(*args[, destination, prefix, ...])Return a dictionary containing references to the whole state of the module.
to
(*args, **kwargs)Move and/or cast the parameters and buffers.
to_empty
(*, device[, recurse])Move the parameters and buffers to the specified device without copying storage.
train
([mode])Set the module in training mode.
type
(dst_type)Casts all parameters and buffers to
dst_type
.xpu
([device])Move all model parameters and buffers to the XPU.
zero_grad
([set_to_none])Reset gradients of all model parameters.
Attributes
T_destination
call_super_init
dump_patches
input_size
hidden_size
bias
dropout
batch_first
training