在Keras的Conv2D和Dense期间，数据如何变化？

Question

正如标题所说。此代码仅适用于：

x = Flatten()(x)

在卷积层和密集层之间。

import numpy as np
import keras
from keras.models import Sequential, Model
from keras.layers import Dense, Dropout, Flatten, Input
from keras.layers import Conv2D, MaxPooling2D
from keras.optimizers import SGD

# Generate dummy data
x_train = np.random.random((100, 100, 100, 3))
y_train = keras.utils.to_categorical(np.random.randint(10, size=(100, 1)), num_classes=10)

#Build Model
input_layer = Input(shape=(100, 100, 3))
x = Conv2D(32, (3, 3), activation='relu')(input_layer)
x = Dense(256, activation='relu')(x)
x = Dense(10, activation='softmax')(x)
model = Model(inputs=[input_layer],outputs=[x])

#compile network
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='categorical_crossentropy', optimizer=sgd)

#train network
model.fit(x_train, y_train, batch_size=32, epochs=10)

否则，我收到此错误：

Traceback (most recent call last):

File "/home/michael/practice_example.py", line 44, in <module>
    model.fit(x_train, y_train, batch_size=32, epochs=10)

File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 1435, in fit
    batch_size=batch_size)

File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 1315, in _standardize_user_data
    exception_prefix='target')

File "/usr/local/lib/python2.7/dist-packages/keras/engine/training.py", line 127, in _standardize_input_data
    str(array.shape))

ValueError: Error when checking target: expected dense_2 to have 4 dimensions, but got array with shape (100, 10)

为什么输出有4个尺寸而没有flatten()图层？

Answer 1

根据keras doc，

  

Conv2D输出形状

     

4D张量与形状:(样本，过滤器，new_rows，new_cols）如果data_format ='channels_first'或4D张量与形状：（samples，new_rows，new_cols，filters）如果data_format ='channels_last'。由于填充，行和列值可能已更改。

由于您使用的是channels_last，因此图层输出的形状为：

# shape=(100, 100, 100, 3)

x = Conv2D(32, (3, 3), activation='relu')(input_layer)
# shape=(100, row, col, 32)

x = Flatten()(x)
# shape=(100, row*col*32)    

x = Dense(256, activation='relu')(x)
# shape=(100, 256)

x = Dense(10, activation='softmax')(x)
# shape=(100, 10)

错误说明（编辑，感谢@Marcin）

使用Dense层将4D张量（shape =（100，row，col，32））连接到2D（张数=（100,256））仍然会形成4D张量（shape =（ 100，row，col，256））这不是你想要的。

# shape=(100, 100, 100, 3)

x = Conv2D(32, (3, 3), activation='relu')(input_layer)
# shape=(100, row, col, 32)

x = Dense(256, activation='relu')(x)
# shape=(100, row, col, 256)

x = Dense(10, activation='softmax')(x)
# shape=(100, row, col, 10)

当输出4D张量与目标2D张量不匹配时，会发生错误。

这就是为什么你需要一个Flatten层来将它从4D平移到2D。

参考

Conv2D Dense

Answer 2

从Dense文档中可以看到，如果Dense的输入具有两个以上的维度 - 它仅应用于最后一个维度 - 并保留所有其他维度：

# shape=(100, 100, 100, 3)

x = Conv2D(32, (3, 3), activation='relu')(input_layer)
# shape=(100, row, col, 32)

x = Dense(256, activation='relu')(x)
# shape=(100, row, col, 256)

x = Dense(10, activation='softmax')(x)
# shape=(100, row, col, 10)

这就是预期4d目标的原因。