sub_modules

Implementation of nn.Modules for temporal fusion transformer.

Classes

AddNorm(input_size, skip_size, trainable_add)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
GateAddNorm(input_size, hidden_size, …)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
GatedLinearUnit(input_size, hidden_size, dropout)	Gated Linear Unit
GatedResidualNetwork(input_size, …)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
InterpretableMultiHeadAttention(n_head, …)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
PositionalEncoder(d_model[, max_seq_len])	Initializes internal Module state, shared by both nn.Module and ScriptModule.
ResampleNorm(input_size, output_size, …)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
ScaledDotProductAttention(dropout, scale)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
TimeDistributed(module, batch_first)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
TimeDistributedInterpolation(output_size, …)	Initializes internal Module state, shared by both nn.Module and ScriptModule.
VariableSelectionNetwork(input_sizes, int], …)	Calcualte weights for num_inputs variables which are each of size input_size

TemporalFusionTransformer AddNorm