对于 Keras LSTM w/return_sequences=True（和 sklearn API），“y 的形状无效”

Question

我有一个序列正在尝试使用带有 return_sequences=True 的 Keras LSTM 进行分类。我有“数据”和“标签”数据集，它们的形状相同 - 二维矩阵，行按位置排列，列按时间间隔排列（单元格值是我的“信号”特征）。因此，带有 return_sequences=True 的 RNN 似乎是一种直观的方法。

将我的数据(X)和标签重塑(Y)为 shape 的 3D 张量后(rows, cols, 1)，我调用model.fit(X, Y)但收到以下错误：

ValueError('y 的形状无效')

它向我指出了类 KerasClassifier() 的 fit 方法的代码，该方法检查len(y.shape)==2.

好吧，也许我应该将我的 2D 重塑'X'为形状（行、列、1）的 3D 张量，但将我的标签保留为 sklearn 界面的 2D？但当我尝试时，我收到另一个 Keras 错误：

ValueError：检查模型目标时出错：预期 lstm_17 有 3 个维度，但得到形状为 (500, 2880) 的数组

...那么如何使用 Sklearn 风格的 Keras RNN 来返回序列呢？Keras 的不同部分似乎要求我的目标是 2D 和 3D。或者（更有可能）我误解了一些东西。

...这是一个可重现的代码示例：

from keras.layers import LSTM
from keras.wrappers.scikit_learn import KerasClassifier

# Raw Data/Targets    
X = np.array([1,2,3,4,5,6,7,8,9,10,11,12]).reshape(3,4)
Y = np.array([1,0,1,1,0,1,0,1,0,1,0,1]).reshape(3,4)

# Convert X to 3D tensor per Keras doc for recurrent layers
X = X.reshape(X.shape[0], X.shape[1], 1)

# .fit() at bottom will throw an error whether or not this line is used to reshape Y
to reshape Y
Y = Y.reshape(Y.shape[0], Y.shape[1], 1)


# Define function to return compiled Keras Model (to pass to Sklearn API)
def keras_rnn(timesteps, num_features):
    '''Function to return compiled Keras Classifier to pass to sklearn wrapper'''

    model = Sequential()
    model.add(LSTM(8, return_sequences=True, input_shape=(timesteps, num_features)))
    model.add(LSTM(1, return_sequences=True, activation = 'sigmoid'))

    model.compile(optimizer = 'RMSprop', loss = 'categorical_crossentropy')
    return model

# Convert compiled Keras model to Scikit-learn-style classifier (compatible w/ sklearn model-tuning methods)
rnn_sklearn = KerasClassifier(build_fn=keras_rnn, 
                        timesteps=4,
                        num_features=1) 

# Fit RNN Model to Data, Target                            
rnn_sklearn.fit(X, Y)

ValueError：y 的形状无效

Answer 1

此代码适用于 Keras 2.0.2：

import numpy as np
from keras.models import Sequential
from keras.layers import LSTM, Flatten
from keras.wrappers.scikit_learn import KerasClassifier

# Raw Data/Targets    
X = np.array([1,2,3,4,5,6,7,8,9,10,11,12]).reshape(3,4)
Y = np.array([1,0,1,1,0,1,0,1,0,1,0,1]).reshape(3,4)

# Convert X to 3D tensor per Keras doc for recurrent layers
X = X.reshape(X.shape[0], X.shape[1], 1)

# .fit() at bottom will throw an error whether or not this line is used to reshape Y to reshape Y
Y = Y.reshape(Y.shape[0], Y.shape[1], 1)


# Define function to return compiled Keras Model (to pass to Sklearn API)
def keras_rnn(timesteps, num_features):
    '''Function to return compiled Keras Classifier to pass to sklearn wrapper'''

    model = Sequential()
    model.add(LSTM(8, return_sequences=True, input_shape=(timesteps, num_features)))
    model.add(LSTM(1, return_sequences=True, activation = 'sigmoid'))

    model.compile(optimizer = 'RMSprop', loss = 'binary_crossentropy')
    return model

# Convert compiled Keras model to Scikit-learn-style classifier (compatible w/ sklearn model-tuning methods)
rnn_sklearn = KerasClassifier(build_fn=keras_rnn, 
                        timesteps=4,
                        num_features=1) 

# Fit RNN Model to Data, Target                            
rnn_sklearn.fit(X, Y)