tf.keras.layers.SimpleRNN

Fully-connected RNN where the output is to be fed back as the new input.

Inherits From: RNN, Layer, Operation

unitsPositive integer, dimensionality of the output space.
activationActivation function to use. Default: hyperbolic tangent (tanh). If you pass None, no activation is applied (ie. "linear" activation: a(x) = x).
use_biasBoolean, (default True), whether the layer uses a bias vector.
kernel_initializerInitializer for the kernel weights matrix, used for the linear transformation of the inputs. Default: "glorot_uniform".
recurrent_initializerInitializer for the recurrent_kernel weights matrix, used for the linear transformation of the recurrent state. Default: "orthogonal".
bias_initializerInitializer for the bias vector. Default: "zeros".
kernel_regularizerRegularizer function applied to the kernel weights matrix. Default: None.
recurrent_regularizerRegularizer function applied to the recurrent_kernel weights matrix. Default: None.
bias_regularizerRegularizer function applied to the bias vector. Default: None.
activity_regularizerRegularizer function applied to the output of the layer (its "activation"). Default: None.
kernel_constraintConstraint function applied to the kernel weights matrix. Default: None.
recurrent_constraintConstraint function applied to the recurrent_kernel weights matrix. Default: None.
bias_constraintConstraint function applied to the bias vector. Default: None.
dropoutFloat between 0 and 1. Fraction of the units to drop for the linear transformation of the inputs. Default: 0.
recurrent_dropoutFloat between 0 and 1. Fraction of the units to drop for the linear transformation of the recurrent state. Default: 0.
return_sequencesBoolean. Whether to return the last output in the output sequence, or the full sequence. Default: False.
return_stateBoolean. Whether to return the last state in addition to the output. Default: False.
go_backwardsBoolean (default: False). If True, process the input sequence backwards and return the reversed sequence.
statefulBoolean (default: False). If True, the last state for each sample at index i in a batch will be used as initial state for the sample of index i in the following batch.
unrollBoolean (default: False). If True, the network will be unrolled, else a symbolic loop will be used. Unrolling can speed-up a RNN, although it tends to be more memory-intensive. Unrolling is only suitable for short sequences.

sequenceA 3D tensor, with shape [batch, timesteps, feature].
maskBinary tensor of shape [batch, timesteps] indicating whether a given timestep should be masked. An individual True entry indicates that the corresponding timestep should be utilized, while a False entry indicates that the corresponding timestep should be ignored.
trainingPython boolean indicating whether the layer should behave in training mode or in inference mode. This argument is passed to the cell when calling it. This is only relevant if dropout or recurrent_dropout is used.
initial_stateList of initial state tensors to be passed to the first call of the cell.

Example:

inputs = np.random.random((32, 10, 8))
simple_rnn = keras.layers.SimpleRNN(4)
output = simple_rnn(inputs)  # The output has shape `(32, 4)`.
simple_rnn = keras.layers.SimpleRNN(
    4, return_sequences=True, return_state=True
)
# whole_sequence_output has shape `(32, 10, 4)`.
# final_state has shape `(32, 4)`.
whole_sequence_output, final_state = simple_rnn(inputs)

activation

bias_constraint

bias_initializer

bias_regularizer

dropout

inputRetrieves the input tensor(s) of a symbolic operation.

Only returns the tensor(s) corresponding to the first time the operation was called.

kernel_constraint

kernel_initializer

kernel_regularizer

outputRetrieves the output tensor(s) of a layer.

Only returns the tensor(s) corresponding to the first time the operation was called.

recurrent_constraint

recurrent_dropout

recurrent_initializer

recurrent_regularizer

units

use_bias

Methods

from_config

View source

Creates a layer from its config.

This method is the reverse of get_config, capable of instantiating the same layer from the config dictionary. It does not handle layer connectivity (handled by Network), nor weights (handled by set_weights).

Args
configA Python dictionary, typically the output of get_config.

Returns
A layer instance.

get_initial_state

View source

inner_loop

View source

reset_state

View source

reset_states

View source

symbolic_call

View source