【问题标题】:Variable batch sizes don't work with tf.keras.layers.RNN when dropout is used (TF2.0)?使用 dropout (TF2.0) 时,可变批量大小不适用于 tf.keras.layers.RNN?
【发布时间】:2020-02-26 06:29:38
【问题描述】:

我想将 RNN 包装器与多个 LSTM 单元一起使用,并带有 dropout。但是,如果批量大小发生变化,我会收到错误消息。

当我删除 dropout 时,代码可以正常工作,所以我认为问题是批次之间没有重置 dropout 掩码。

import numpy as np
import tensorflow as tf

input_dim = 3
output_dim = 3
num_timesteps = 2
neurons = [32,32]

# Model
input_layer = tf.keras.Input(shape=(num_timesteps, input_dim))
cell = [tf.keras.layers.LSTMCell(n,dropout=.2) for n in neurons]
lstm = tf.keras.layers.RNN(cell,return_state=True,return_sequences=True)
lstm_out, hidden_state, cell_state = lstm(input_layer)
output = tf.keras.layers.Dense(output_dim)(lstm_out)

mdl = tf.keras.Model(
    inputs=input_layer,
    outputs=[hidden_state, cell_state, output]
)

# Run batches of different sizes
batch_1 = np.random.rand(10, num_timesteps, input_dim).astype(np.float32)
h_state, c_state, out = mdl(batch_1) # batch size is 10x2x3

batch_2 = np.random.rand(9, num_timesteps, input_dim).astype(np.float32)
h_state, c_state, out = mdl(batch_2) # batch size is 9x2x3

此代码给出错误:InvalidArgumentError: Incompatible shapes: [9,3] vs. [10,3] [Op:Mul] name: model/rnn/mul/

如果我去掉 dropout,代码就可以工作。我可以以某种方式使用 reset_dropout_mask 吗?它似乎没有被调用。

【问题讨论】:

    标签: python lstm recurrent-neural-network tensorflow2.0 dropout


    【解决方案1】:

    我可以使用Tensorflow Version 2.0.0 重现您的错误。

    但是,如果我将 Tensorflow Version 升级到 2.12.2 并运行相同的代码,则不会出现错误。

    完整的工作代码如下所示:

    !pip install tensorflow==2.2
    
    import numpy as np
    import tensorflow as tf
    
    print(tf.__version__) # Printing the Tensorflow Version just to be Sure
    
    input_dim = 3
    output_dim = 3
    num_timesteps = 2
    neurons = [32,32]
    
    # Model
    input_layer = tf.keras.Input(shape=(num_timesteps, input_dim))
    cell = [tf.keras.layers.LSTMCell(n,dropout=.2) for n in neurons]
    lstm = tf.keras.layers.RNN(cell,return_state=True,return_sequences=True)
    lstm_out, hidden_state, cell_state = lstm(input_layer)
    output = tf.keras.layers.Dense(output_dim)(lstm_out)
    
    mdl = tf.keras.Model(
        inputs=input_layer,
        outputs=[hidden_state, cell_state, output]
    )
    
    # Run batches of different sizes
    batch_1 = np.random.rand(10, num_timesteps, input_dim).astype(np.float32)
    h_state, c_state, out = mdl(batch_1) # batch size is 10x2x3
    
    batch_2 = np.random.rand(9, num_timesteps, input_dim).astype(np.float32)
    h_state, c_state, out = mdl(batch_2) # batch size is 9x2x3
    

    以上代码的输出如下所示:

    2.2.0
    

    希望这会有所帮助。快乐学习!

    【讨论】:

      猜你喜欢
      • 2016-02-26
      • 2016-08-28
      • 2016-12-24
      • 2019-02-02
      • 2019-11-13
      • 2018-07-15
      • 2016-01-25
      • 2020-05-04
      • 2017-12-22
      相关资源
      最近更新 更多