How to use earlystopping in deepCTR-Torch ? 能用earlystopping吗,咋用？

[CFZhai]

An detailed example will be appreciated!

[CFZhai]

自问自答。

[CFZhai]

just add a TF instance of EarlyStopping.

early_stopping = tf.keras.callbacks.EarlyStopping( monitor="val_mse", patience=10, ) #restore_best_weights=True

model.fit(train_model_input,train[target].values,batch_size=32,epochs=1000,verbose=2,validation_split=0.1, callbacks= [early_stopping])

[zanshuxun]

https://deepctr-torch.readthedocs.io/en/latest/FAQ.html#set-learning-rate-and-use-earlystopping