Gfyuki

I can't understand the meaning of "nb_steps_warmup", a parameter of the __init__ function of DQNAgent class of the Keras_RL module.

I just know that when I set small value for "nb_steps_warmup" the command line prints: UserWarning: Not enough entries to sample without replacement. Consider increasing your warm-up phase to avoid oversampling!

This is my code:

import numpy as np

import gym

import gym_briscola

import argparse

import os



from keras.models import Sequential

from keras.layers import Dense, Activation, Flatten

from keras.optimizers import Adam



from rl.agents.dqn import DQNAgent

from rl.policy import BoltzmannQPolicy

from rl.memory import SequentialMemory

import logging





def getModel(input_shape, nb_actions):



    # Next, we build a very simple model.

    model = Sequential()

    model.add(Flatten(input_shape = input_shape))

    model.add(Dense(nb_actions, activation = 'relu'))

    for i in range(2):

        model.add(Dense(2, activation = 'relu'))



    model.add(Dense(nb_actions, activation = 'relu'))

    # print(model.summary())

    return model



def init():



    ENV_NAME = 'Briscola-v0'



    # Get the environment and extract the number of actions.

    env = gym.make(ENV_NAME)

    env.setName("Inteligence")

    env.cicle = True

    nb_actions = env.action_space.n

    window_length = 10



    input_shape = (window_length, 5)



    # Finally, we configure and compile our agent. You can use every built-in Keras optimizer and

    # even the metrics!

    memory = SequentialMemory(limit=50000, window_length=window_length)



    #Maxwell-boltzmann distribution

    policy = BoltzmannQPolicy()

    dqn = DQNAgent(model=getModel(input_shape, nb_actions), nb_actions=nb_actions, memory=memory, nb_steps_warmup=11,

                   target_model_update=1e-5, policy=policy, dueling_type='avg', enable_dueling_network=True)



    print("Compila")

    dqn.compile(Adam(lr=1e-5), metrics=['mae'])



    try:

        dqn.load_weights('dqn_{}_weights.h5f'.format(ENV_NAME))

    except OSError:

        print("File non trovato")



    return dqn, env, ENV_NAME



def startTraining():



    dqn, env, ENV_NAME = init()



    print("Fit")

    dqn.fit(env, nb_steps=5E6, visualize=False, verbose=1, log_interval=1000)



    # After training is done, we save the final weights.

    dqn.save_weights('dqn_{}_weights.h5f'.format(ENV_NAME), overwrite=True)





def startTest():



    dqn, env, _ = init()



    print("Test")

    # Finally, evaluate our algorithm for 5 episodes.

    dqn.test(env, nb_episodes=10, visualize=False)



#Log config

def setLogging(show = True):

    logging.getLogger("Briscola").propagate = show

    logging.getLogger("IA").propagate = True

    logging.getLogger("Client").propagate = show

    logging.getLogger("Vincitore").propagate = show



    logging.basicConfig(level=logging.INFO)



if __name__ == "__main__":



    #Parameter settings

    parser = argparse.ArgumentParser()

    parser.add_argument("-m", "--modality", help="The modality of the program", const="train", nargs='?')

    parser.add_argument("-l", "--logging", help="Enable logging", type=bool, const=False, nargs='?')

    args = parser.parse_args()



    setLogging(True)

    print("Avvio modalita' ", args.modality)



    if args.modality == "test":

        startTest()

    else:

        startTraining()



    print("Fine")

This is the doc of the module: https://keras-rl.readthedocs.io/en/latest/agents/dqn/

I hope I was clear with my English.