Class GRU

Gated Recurrent Unit - Cho et al. 2014. There are two variants.The default one is based on 1406.1078v3 and has reset gate applied to hidden state before matrix multiplication. The other one is based on original 1406.1078v1 and has the order reversed. The second variant is compatible with CuDNNGRU (GPU-only) and allows inference on CPU.Thus it has separate biases for kernel and recurrent_kernel.Use 'reset_after'=True and recurrent_activation='sigmoid'.

Inheritance

System.Object

Keras

Base

BaseLayer

RNN

GRU

Implements

System.IDisposable

Inherited Members

BaseLayer.Set(BaseLayer[])

Base.Parameters

Base.None

Base.Init()

Base.ToPython()

Base.InvokeStaticMethod(Object, String, Dictionary<String, Object>)

Base.InvokeMethod(String, Dictionary<String, Object>)

Base.Item[String]

Keras.Instance

Keras.keras

Keras.keras2onnx

Keras.tfjs

Keras.Dispose()

Keras.ToTuple(Array)

Keras.ToList(Array)

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: Keras.Layers

Assembly: Keras.dll

Syntax

public class GRU : RNN, IDisposable

Constructors

| Improve this Doc View Source

GRU(Int32, String, String, Boolean, String, String, String, String, String, String, String, String, String, String, Single, Single, Int32, Boolean, Boolean, Boolean, Boolean, Boolean, Boolean)

Initializes a new instance of the GRU class.

Declaration

public GRU(int units, string activation = "tanh", string recurrent_activation = "hard_sigmoid", bool use_bias = true, string kernel_initializer = "glorot_uniform", string recurrent_initializer = "orthogonal", string bias_initializer = "zeros", string kernel_regularizer = "", string recurrent_regularizer = "", string bias_regularizer = "", string activity_regularizer = "", string kernel_constraint = "", string recurrent_constraint = "", string bias_constraint = "", float dropout = 0F, float recurrent_dropout = 0F, int implementation = 1, bool return_sequences = false, bool return_state = false, bool go_backwards = false, bool stateful = false, bool unroll = false, bool reset_after = false)

Parameters

Type	Name	Description
System.Int32	units	Positive integer, dimensionality of the output space.
System.String	activation	Activation function to use (see activations). Default: hyperbolic tangent (tanh). If you pass None, no activation is applied (ie. "linear" activation: a(x) = x).
System.String	recurrent_activation	Activation function to use for the recurrent step (see activations). Default: hard sigmoid (hard_sigmoid). If you pass None, no activation is applied (ie. "linear" activation: a(x) = x).
System.Boolean	use_bias	Boolean, whether the layer uses a bias vector.
System.String	kernel_initializer	Initializer for the kernel weights matrix, used for the linear transformation of the inputs (see initializers).
System.String	recurrent_initializer	Initializer for the recurrent_kernel weights matrix, used for the linear transformation of the recurrent state (see initializers).
System.String	bias_initializer	Initializer for the bias vector (see initializers).
System.String	kernel_regularizer	Regularizer function applied to the kernel weights matrix (see regularizer).
System.String	recurrent_regularizer	Regularizer function applied to the recurrent_kernel weights matrix (see regularizer).
System.String	bias_regularizer	Regularizer function applied to the bias vector (see regularizer).
System.String	activity_regularizer	Regularizer function applied to the output of the layer (its "activation"). (see regularizer).
System.String	kernel_constraint	Constraint function applied to the kernel weights matrix (see constraints).
System.String	recurrent_constraint	Constraint function applied to the recurrent_kernel weights matrix (see constraints).
System.String	bias_constraint	Constraint function applied to the bias vector (see constraints).
System.Single	dropout	Float between 0 and 1. Fraction of the units to drop for the linear transformation of the inputs.
System.Single	recurrent_dropout	Float between 0 and 1. Fraction of the units to drop for the linear transformation of the recurrent state.
System.Int32	implementation	Implementation mode, either 1 or 2. Mode 1 will structure its operations as a larger number of smaller dot products and additions, whereas mode 2 will batch them into fewer, larger operations. These modes will have different performance profiles on different hardware and for different applications.
System.Boolean	return_sequences	Boolean. Whether to return the last output in the output sequence, or the full sequence.
System.Boolean	return_state	Boolean. Whether to return the last state in addition to the output.
System.Boolean	go_backwards	Boolean (default False). If True, process the input sequence backwards and return the reversed sequence.
System.Boolean	stateful	Boolean (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.
System.Boolean	unroll	Boolean (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.
System.Boolean	reset_after	GRU convention (whether to apply reset gate after or before matrix multiplication). False = "before" (default), True = "after" (CuDNN compatible).

Implements

System.IDisposable