We can load the models we saved in the training script with keras.models.load_model. I created this helper to do just that. We will load the model using the following code:
def load_models(): model = load_model('char_s2s.h5') encoder_model = load_model('char_s2s_encoder.h5') decoder_model = load_model('char_s2s_decoder.h5') return [model, encoder_model, decoder_model]
We can just call the following function to load all the three models:
model, encoder_model, decoder_model = load_models()