一尘不染

Python-Keras,如何获得每一层的输出?

python

我已经使用CNN训练了二进制分类模型,这是我的代码

model = Sequential()
model.add(Convolution2D(nb_filters, kernel_size[0], kernel_size[1],
                        border_mode='valid',
                        input_shape=input_shape))
model.add(Activation('relu'))
model.add(Convolution2D(nb_filters, kernel_size[0], kernel_size[1]))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=pool_size))
# (16, 16, 32)
model.add(Convolution2D(nb_filters*2, kernel_size[0], kernel_size[1]))
model.add(Activation('relu'))
model.add(Convolution2D(nb_filters*2, kernel_size[0], kernel_size[1]))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=pool_size))
# (8, 8, 64) = (2048)
model.add(Flatten())
model.add(Dense(1024))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(2))  # define a binary classification problem
model.add(Activation('softmax'))

model.compile(loss='categorical_crossentropy',
              optimizer='adadelta',
              metrics=['accuracy'])
model.fit(x_train, y_train,
          batch_size=batch_size,
          nb_epoch=nb_epoch,
          verbose=1,
          validation_data=(x_test, y_test))

在这里,我想像TensorFlow一样获得每一层的输出,我该怎么做?


阅读 1237

收藏
2020-02-17

共1个答案

一尘不染

你可以使用以下命令轻松获取任何图层的输出: model.layers[index].output

对于所有图层,请使用以下命令:

from keras import backend as K

inp = model.input                                           # input placeholder
outputs = [layer.output for layer in model.layers]          # all layer outputs
functors = [K.function([inp, K.learning_phase()], [out]) for out in outputs]    # evaluation functions

# Testing
test = np.random.random(input_shape)[np.newaxis,...]
layer_outs = [func([test, 1.]) for func in functors]
print layer_outs

注:为了模拟差使用learning_phase如1.在layer_outs以其它方式使用0.

编辑:(基于评论)

K.function 创建theano / tensorflow张量函数,该函数随后用于从给定输入的符号图中获取输出。

现在K.learning_phase()需要输入作为输入,因为许多Keras层(如Dropout / Batchnomalization)都依赖它来在训练和测试期间更改行为。

因此,如果你删除代码中的辍学层,则可以简单地使用:

from keras import backend as K

inp = model.input                                           # input placeholder
outputs = [layer.output for layer in model.layers]          # all layer outputs
functors = [K.function([inp], [out]) for out in outputs]    # evaluation functions

# Testing
test = np.random.random(input_shape)[np.newaxis,...]
layer_outs = [func([test]) for func in functors]
print layer_outs

编辑2:更优化

我刚刚意识到,先前的答案并不是针对每个函数评估进行优化的,因为数据将被传输到CPU-> GPU内存中,并且还需要对低层进行n-n-over的张量计算。

相反,这是一种更好的方法,因为你不需要多个函数,而只需一个函数即可为你提供所有输出的列表:

from keras import backend as K

inp = model.input                                           # input placeholder
outputs = [layer.output for layer in model.layers]          # all layer outputs
functor = K.function([inp, K.learning_phase()], outputs )   # evaluation function

# Testing
test = np.random.random(input_shape)[np.newaxis,...]
layer_outs = functor([test, 1.])
print layer_outs
2020-02-17