Question

现在我在这里使用早期停留在Keras：

X,y= load_data('train_data')
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=12)

datagen = ImageDataGenerator(
    horizontal_flip=True,
    vertical_flip=True)

early_stopping_callback = EarlyStopping(monitor='val_loss', patience=epochs_to_wait_for_improve)
history = model.fit_generator(datagen.flow(X_train, y_train, batch_size=batch_size),
            steps_per_epoch=len(X_train) / batch_size, validation_data=(X_test, y_test),
            epochs=n_epochs, callbacks=[early_stopping_callback])

但在model.fit_generator结束时，它会在epochs_to_wait_for_improve之后保存模型，但我想用min val_loss保存模型是否有意义并且可能吗？

Answer 1

是的，可以再一次回调，这是代码：

early_stopping_callback = EarlyStopping(monitor='val_loss', patience=epochs_to_wait_for_improve)
checkpoint_callback = ModelCheckpoint(model_name+'.h5', monitor='val_loss', verbose=1, save_best_only=True, mode='min')
history = model.fit_generator(datagen.flow(X_train, y_train, batch_size=batch_size),
            steps_per_epoch=len(X_train) / batch_size, validation_data=(X_test, y_test),
            epochs=n_epochs, callbacks=[early_stopping_callback, checkpoint_callback])

Keras：提前停止模型保存

1 个答案: