如何在keras中记录val_loss和损失预批量

Question

我在keras中使用回调函数来记录loss和val_loss每个时期，但是我希望每个批次都做同样的工作。我发现了一个名为on_batch_begin(self,batch,log={})的回调函数，但我不确定如何使用它。

Answer 1

继和here modyfing一个例子：

class LossHistory(keras.callbacks.Callback): 
    def on_train_begin(self, logs={}): 
     self.losses = [] 
     self.val_losses = [] 


def on_batch_end(self, batch, logs={}): 
    self.losses.append(logs.get('loss')) 
    self.val_losses.append(logs.get('val_loss')) 

model = Sequential() 
model.add(Dense(10, input_dim=784, init='uniform')) 
model.add(Activation('softmax')) 
model.compile(loss='categorical_crossentropy', optimizer='rmsprop') 

history = LossHistory() 
model.fit(X_train, Y_train, batch_size=128, nb_epoch=20, verbose=0, validation_split=0.1, 
      callbacks=[history]) 

print history.losses 
# outputs 
''' 
[0.66047596406559383, 0.3547245744908703, ..., 0.25953155204159617, 0.25901699725311789] 
''' 
print history.val_losses