多任务 CNN model 提前停止，带有验证损失

Question

I am trying to fit a multitasking model with validation data and tracing validation loss for early stopping. Is there any way to trace and early stop with validation loss? My demo code is following it shows a warning that validation loss is not available.

    def main_model(height, width): 
            input_img = Input(shape = (height, width, 1))
            conv01_1 = Conv2D(64, (3, 3), activation='relu', padding='same')(input_img)
            pool01_1 = AveragePooling2D(pool_size=(2, 2),strides=None, padding="same")( conv01_1)
            batch_nor01_1= BatchNormalization()(pool01_1)
            drout01_1= Dropout(0.1)(batch_nor01_1)
            flatten_layer = Flatten()(drout01_1)
            x1_dense = Dense(512,activation='relu')( flatten_layer )
            out_1=Dense(6,activation = "softmax",name='activity_output')( x1_dense)
            out_2=Dense(1,activation='linear',name='energy_output')( x1_dense)
            model = Model(inputs=input_img,outputs = [out_1,out_2])
            model.compile(optimizer=SGD(lr=0.001,momentum=0.9),loss={'activity_output':'categorical_crossentropy', 'energy_output': 'mse'},loss_weights={'activity_output': 0.5, 'energy_output': 0.5},metrics=['accuracy','mae'])
            model.summary()
            return model
    
    model_name=s+'_best_model.h5'
    mc = ModelCheckpoint(model_name, monitor='validation_loss', mode='auto', verbose=1, save_best_only=True)
    es = EarlyStopping(monitor='validation_loss',min_delta=0,patience=20,verbose=0, mode='auto')
    ```
    
    batch_size=500
    epochs=1
    model=main_model(height, width)
    History = model.fit(x_train,[y_train,y_train_1],epochs = epochs, validation_data = (x_valid,y_valid,y_valid_1),verbose = 1,callbacks=[callback_test,es,lrs,mc,])

'''


        

Answer 1

I have got the solution. Basically, I have replaced the validation_loss with val_loss so the code is now:

mc = ModelCheckpoint(model_name, monitor='val_loss', mode='min', verbose=1, 
                     save_best_only=True)