python - 多条件提前停止

Question

我正在为推荐系统(项目推荐)进行多类分类，我目前正在使用 sparse_categorical_crossentropy 训练我的网络损失。因此，执行EarlyStopping是合理的。通过监控我的验证损失，val_loss像这样:

tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=10)

它按预期工作。然而，网络(推荐系统)的性能是通过 Average-Precision-at-10 来衡量的，并在训练期间作为一个指标进行跟踪，如 average_precision_at_k10 .因此，我还可以使用此指标执行提前停止:

tf.keras.callbacks.EarlyStopping(monitor='average_precision_at_k10', patience=10)

这也按预期工作。
我的问题:
有时验证损失会增加，而 Average-Precision-at-10 会提高，反之亦然。因此，我需要监控 两者 ，并执行提前停止， 当且仅当 两者都在恶化。我想做什么:

tf.keras.callbacks.EarlyStopping(monitor=['val_loss', 'average_precision_at_k10'], patience=10)

这显然不起作用。任何想法如何做到这一点？

Answer 1

在 Gerry P 的指导下上面我设法创建了自己的自定义 EarlyStopping 回调，并认为我将其发布在这里以防其他人希望实现类似的功能。
如 两者 验证损失 和 patience 的平均精度在 10 处没有提高epoch 数，执行提前停止。

class CustomEarlyStopping(keras.callbacks.Callback):
    def __init__(self, patience=0):
        super(CustomEarlyStopping, self).__init__()
        self.patience = patience
        self.best_weights = None
        
    def on_train_begin(self, logs=None):
        # The number of epoch it has waited when loss is no longer minimum.
        self.wait = 0
        # The epoch the training stops at.
        self.stopped_epoch = 0
        # Initialize the best as infinity.
        self.best_v_loss = np.Inf
        self.best_map10 = 0

    def on_epoch_end(self, epoch, logs=None): 
        v_loss=logs.get('val_loss')
        map10=logs.get('val_average_precision_at_k10')

        # If BOTH the validation loss AND map10 does not improve for 'patience' epochs, stop training early.
        if np.less(v_loss, self.best_v_loss) and np.greater(map10, self.best_map10):
            self.best_v_loss = v_loss
            self.best_map10 = map10
            self.wait = 0
            # Record the best weights if current results is better (less).
            self.best_weights = self.model.get_weights()
        else:
            self.wait += 1
            if self.wait >= self.patience:
                self.stopped_epoch = epoch
                self.model.stop_training = True
                print("Restoring model weights from the end of the best epoch.")
                self.model.set_weights(self.best_weights)
                
    def on_train_end(self, logs=None):
        if self.stopped_epoch > 0:
            print("Epoch %05d: early stopping" % (self.stopped_epoch + 1))

然后用作:

model.fit(
    x_train,
    y_train,
    batch_size=64,
    steps_per_epoch=5,
    epochs=30,
    verbose=0,
    callbacks=[CustomEarlyStopping(patience=10)],
)