DynamicGRUV2 Constructor

Description

Used for RNN training and inference with TensorFlow.

Prototype

class DynamicGRUV2(_DynamicBasic):
def __init__(self,
hidden_size,
dtype,
direction=DYNAMIC_RNN_UNIDIRECTION,
cell_depth=1,
keep_prob=1.0,
cell_clip=-1.0,
num_proj=0,
time_major=True,
activation="tanh",
gate_order="zrh",
reset_after=True,
is_training=True)

Options

Option	Input/Output	Description
hidden_size	Input	Dimension of the GRU hidden state h.
dtype	Input	Data type of initialized weight and bias. Note: The input data will be converted into the type supported by DynamicRNN.
direction	Input	(Optional) Only DYNAMIC_RNN_UNIDIRECTION is supported.
cell_depth	Input	(Optional) Only one layer is supported.
keep_prob	Input	(Optional) Dropout is not supported.
cell_clip	Input	(Optional) Cell state clipping is not supported.
num_proj	Input	(Optional) Projection calculation is not supported.
time_major	Input	(Optional) Only input x in [num_step, batch_size, embedding] is supported.
activation	Input	(Optional) Only "tanh" is supported.
gate_order	Input	(Optional) Gate order. Defaults to "zrh". "rzh" is also used commonly.
reset_after	Input	(Optional) Defaults to TRUE, indicating that the reset gate is applied to the hidden state after matrix multiplication.
is_training	Input	(Optional) By default, the training mode is used.

Returns

output_y: RNN output tensor, with the shape of [num_step, batch_size, hidden_size].

output_h: RNN output tensor, with the shape of [num_step, batch_size, hidden_size].

update: RNN intermediate result, which is used for backward propagation.

reset: RNN intermediate result, which is used for backward propagation.

new: RNN intermediate result, which is used for backward propagation.

hidden_new: RNN intermediate result, which is used for backward propagation.

Restrictions

Currently, there are restrictions on the API. See the following table for details.

Example

import tensorflow as tf
from npu_bridge.estimator.npu.npu_dynamic_rnn import DynamicGRUV2

inputs = tf.random.normal(shape=(25, 64, 128))
gru = DynamicGRUV2(hidden_size=16, dtype=tf.float16, is_training=False)
y, output_h, update, reset, new, hidden_new = gru(inputs, seq_length=None, init_h=None)

Parent topic: npu_bridge.estimator.npu.npu_dynamic_rnn