下面是model.fit_generator()函数的一些参数。这些对象保存在标有回调的列表中。
checkpoint = ModelCheckpoint(
model_file,
monitor= 'val_acc',
save_best_only=True)
early_stopping = EarlyStopping(
monitor='val_loss',
patience=5,
verbose=1,
restore_best_weights=True)
tensorboard = TensorBoard(
log_dir=log_dir,
batch_size=batch_size,
update_freq = 'batch')
reduce_lr = ReduceLROnPlateau(
monitor='val_loss',
patience=5,
cooldown=2,
min_lr=0.0000000001,
verbose=1)
#-----------------------------------------------------------------------------------------------------------------#
callbacks = [checkpoint, reduce_lr, early_stopping, tensorboard]
创建了回调对象和对象的参数后,我实现了层并进行了编译(未显示,因为它与我遇到的问题无关)。然后运行model.fit_generator函数(使用上面的回调参数):
history = model.fit_generator(
train_generator,
steps_per_epoch = steps_per_epoch,
epochs=epochs,
verbose=2,
callbacks=callbacks,
validation_data=validation_generator,
validation_steps=validation_steps,
class_weight=class_weight)
我得到的错误是:
KeyError: 'val_acc'
据我了解,这意味着val_acc不在列表中。但这是..因此需要帮助来了解为什么我会收到此错误。
根据有关keras中回调的官方文档:
checkpoint = ModelCheckpoint(
model_file,
monitor= 'val_acc',
save_best_only=True)
更改为:
checkpoint = ModelCheckpoint(
model_file,
monitor= 'val_loss',
save_best_only=True)
供参考,请访问:https://keras.io/callbacks/