GRUCell¶

class paddle.fluid.layers. GRUCell ( hidden_size, param_attr=None, bias_attr=None, gate_activation=None, activation=None, dtype='float32', name='GRUCell' ) [source]

api_attr

Static Graph

Gated Recurrent Unit cell. It is a wrapper for fluid.contrib.layers.rnn_impl.BasicGRUUnit to make it adapt to RNNCell.

The formula used is as follow:

\[ \begin{align}\begin{aligned}u_t & = act_g(W_{ux}x_{t} + W_{uh}h_{t-1} + b_u)\\r_t & = act_g(W_{rx}x_{t} + W_{rh}h_{t-1} + b_r)\\\begin{split}\\tilde{h_t} & = act_c(W_{cx}x_{t} + W_{ch}(r_t \odot h_{t-1}) + b_c)\end{split}\\\begin{split}h_t & = u_t \odot h_{t-1} + (1-u_t) \odot \\tilde{h_t}\end{split}\end{aligned}\end{align} \]

For more details, please refer to Learning Phrase Representations using RNN Encoder Decoder for Statistical Machine Translation

Examples

import paddle.fluid.layers as layers
cell = layers.GRUCell(hidden_size=256)

call ( inputs, states )

call¶

Perform calculations of GRU.

Parameters

• inputs (Variable) – A tensor with shape [batch_size, input_size], corresponding to \(x_t\) in the formula. The data type should be float32 or float64.

• states (Variable) – A tensor with shape [batch_size, hidden_size]. corresponding to \(h_{t-1}\) in the formula. The data type should be float32 or float64.

Returns

A tuple( (outputs, new_states) ), where outputs and

new_states is the same tensor shaped [batch_size, hidden_size], corresponding to \(h_t\) in the formula. The data type of the tensor is same as that of states.

Return type

tuple

property state_shape

The state_shape of GRUCell is a shape [hidden_size] (-1 for batch size would be automatically inserted into shape). The shape corresponds to \(h_{t-1}\).

get_initial_states ( batch_ref, shape=None, dtype='float32', init_value=0, batch_dim_idx=0 )

get_initial_states¶

Generate initialized states according to provided shape, data type and value.

Parameters

• batch_ref – A (possibly nested structure of) tensor variable[s]. The first dimension of the tensor will be used as batch size to initialize states.

• shape – A (possibly nested structure of) shape[s], where a shape is represented as a list/tuple of integer). -1(for batch size) will beautomatically inserted if shape is not started with it. If None, property state_shape will be used. The default value is None.

• dtype – A (possibly nested structure of) data type[s]. The structure must be same as that of shape, except when all tensors’ in states has the same data type, a single data type can be used. If property cell.state_shape is not available, float32 will be used as the data type. The default value is float32.

• init_value – A float value used to initialize states.

• batch_dim_idx – An integer indicating which dimension of the tensor in inputs represents batch size. The default value is 0.

Returns

tensor variable[s] packed in the same structure provided

by shape, representing the initialized states.

Return type

Variable

property state_dtype

Abstract method (property). Used to initialize states. A (possibly nested structure of) data types[s]. The structure must be same as that of shape, except when all tensors’ in states has the same data type, a single data type can be used. Not necessary to be implemented if states are not initialized by get_initial_states or the dtype argument is provided when using get_initial_states.