I have a general question about Keras. When training a Artificial Neural Network (e.g. a Multi-Layer-Perceptron or a LSTM) with a split of training, validation and test data (e.g. 70 %, 20 %, 10 %), I would like to know which parameter configuration the trained model is eventually using for predictions?

Here I have an exmaple from a training process with 11 epoch:

I could think about 3 possible parameter configurations (surely there are also others):

1. The configuration that led to the lowest error in the training dataset (which would be after the 11th epoch)
2. The configuration after the last epoch (which would the after the 11th epoch, as in 1.)
3. The configuration that led to the lowest error in the validation dataset (which would be after the 3rd epoch)

If you just build the model without for example like this:

# Build the model and train it

optimizer_adam = tf.keras.optimizers.Adam(lr= 0.001)
model = keras.models.Sequential([
    keras.layers.LSTM(10, return_sequences=True, input_shape=[None, numberOfInputFeatures]),
    keras.layers.LSTM(10, return_sequences=True),
    keras.layers.TimeDistributed(keras.layers.Dense(numberOfOutputNeurons))
    ])
    
    
model.compile(loss="mean_squared_error", optimizer=optimizer_adam, metrics=['mean_absolute_percentage_error'])
history = model.fit(X_train, Y_train, epochs=11, batch_size=10, validation_data=(X_valid, Y_valid))

# Predict the values from the test dataset
Y_pred = model.predict(X_test)

Can you tell me which configuration is used for predicting the values from the test dataset in the line Y_pred = model.predict(X_test)?

CodePudding user response：

It would be the configuration after the last epoch (the 2nd possible configuration that you have mentioned).