Python tf.compat.v1.nn.dynamic_rnn用法及代碼示例

創建由 RNNCell cell 指定的循環神經網絡。 (已棄用)

用法

tf.compat.v1.nn.dynamic_rnn(
    cell, inputs, sequence_length=None, initial_state=None, dtype=None,
    parallel_iterations=None, swap_memory=False, time_major=False, scope=None
)

遷移到 TF2

警告：這個 API 是為 TensorFlow v1 設計的。繼續閱讀有關如何從該 API 遷移到本機 TensorFlow v2 等效項的詳細信息。見TensorFlow v1 到 TensorFlow v2 遷移指南有關如何遷移其餘代碼的說明。

tf.compat.v1.nn.dynamic_rnn 與即刻執行和 tf.function 不兼容。請改用tf.keras.layers.RNN 進行 TF2 遷移。以 LSTM 為例，您可以使用 tf.keras.layers.LSTMCell 或直接通過 tf.keras.layers.LSTM 實例化 tf.keras.layers.RNN 層。創建 keras 層後，您可以通過調用具有輸入和狀態的層來獲取輸出和狀態。有關 Keras RNN 的更多詳細信息，請參閱本指南。您還可以在本文檔中找到更多關於 Keras RNN 和 TF compat v1 rnn 的區別和比較的詳細信息

到原生 TF2 的結構映射

前：

# create 2 LSTMCells
rnn_layers = [tf.compat.v1.nn.rnn_cell.LSTMCell(size) for size in [128, 256]]

# create a RNN cell composed sequentially of a number of RNNCells
multi_rnn_cell = tf.compat.v1.nn.rnn_cell.MultiRNNCell(rnn_layers)

# 'outputs' is a tensor of shape [batch_size, max_time, 256]
# 'state' is a N-tuple where N is the number of LSTMCells containing a
# tf.nn.rnn_cell.LSTMStateTuple for each cell
outputs, state = tf.compat.v1.nn.dynamic_rnn(cell=multi_rnn_cell,
                                             inputs=data,
                                             dtype=tf.float32)

後：

# RNN layer can take a list of cells, which will then stack them together.
# By default, keras RNN will only return the last timestep output and will not
# return states. If you need whole time sequence output as well as the states,
# you can set `return_sequences` and `return_state` to True.
rnn_layer = tf.keras.layers.RNN([tf.keras.layers.LSTMCell(128),
                                 tf.keras.layers.LSTMCell(256)],
                                return_sequences=True,
                                return_state=True)
outputs, output_states = rnn_layer(inputs, states)

如何映射參數

TF1 參數名稱	TF2 參數名稱	注意
cell	cell	在RNN層構造函數中
inputs	inputs	在RNN層__call__
sequence_length	未使用	在 RNN 之前添加遮罩層：以達到相同的結果。
initial_state	initial_state	在RNN層__call__
dtype	dtype	在RNN層構造函數中
parallel_iterations	不支持
swap_memory	不支持
time_major	time_major	在RNN層構造函數中
scope	不支持

警告：此函數已棄用。它將在未來的版本中刪除。更新說明：請使用keras.layers.RNN(cell)，相當於這個API

執行 inputs 的完全動態展開。

例子：

# create a BasicRNNCell
rnn_cell = tf.compat.v1.nn.rnn_cell.BasicRNNCell(hidden_size)

# 'outputs' is a tensor of shape [batch_size, max_time, cell_state_size]

# defining initial state
initial_state = rnn_cell.zero_state(batch_size, dtype=tf.float32)

# 'state' is a tensor of shape [batch_size, cell_state_size]
outputs, state = tf.compat.v1.nn.dynamic_rnn(rnn_cell, input_data,
                                   initial_state=initial_state,
                                   dtype=tf.float32)

# create 2 LSTMCells
rnn_layers = [tf.compat.v1.nn.rnn_cell.LSTMCell(size) for size in [128, 256]]

# create a RNN cell composed sequentially of a number of RNNCells
multi_rnn_cell = tf.compat.v1.nn.rnn_cell.MultiRNNCell(rnn_layers)

# 'outputs' is a tensor of shape [batch_size, max_time, 256]
# 'state' is a N-tuple where N is the number of LSTMCells containing a
# tf.nn.rnn_cell.LSTMStateTuple for each cell
outputs, state = tf.compat.v1.nn.dynamic_rnn(cell=multi_rnn_cell,
                                   inputs=data,
                                   dtype=tf.float32)