How can i save a trained reinforcement learning agent to avoid training it each time?

Question

I tried using pickle to save the tained agent

   try:
      agent1 = pickle.load(open(model_file_path, 'rb'))
    except:
      print("An exception occurred")
      train_agent(True)

    if agent1 == None:
        train_agent(True)

    human = Human()
    human.set_sym(env1.o)

    agent1.set_verbose(True)
    start_session(agent1, human, Environment(), draw=2)

    pickle.dump(agent1, open(model_file_path, 'wb')) 

    return agent1.prediction

But the file that saves the agent becomes very heavy approx 1gb, hence i am not able to resue the agent

Answer 1

HDF5 Format is a grid format that is ideal for storing multi-dimensional arrays of numbers. For example: using Keras/Tensorflow you can very easy save/load model and weights:

# Save the model
model.save('path_to_my_model.h5')

# Recreate the exact same model purely from the file
new_model = keras.models.load_model('path_to_my_model.h5')

# Save weights
model.save_weights('path_to_my_weights.h5')

# Load weights
new_model.load_weights('path_to_my_weights.h5')