from keras.optimizers import Adam
import os
import sys
sys.path.append(os.path.join(os.path.dirname(__file__), '../'))
from src.r2d2 import R2D2, Actor
from src.r2d2_callbacks import *
from src.processor import AtariProcessor
from src.image_model import DQNImageModel
from src.memory import *
from src.policy import *
from src.common import InputType, LstmType, DuelingNetwork, seed_everything, LoggerType
from src.callbacks import ConvLayerView, MovieLogger
seed_everything(42)
ENV_NAME = "BreakoutDeterministic-v4"
class MyActor(Actor):
def getPolicy(self, actor_index, actor_num):
return EpsilonGreedy(0.1)
def fit(self, index, agent):
env = gym.make(ENV_NAME)
agent.fit(env, visualize=False, verbose=0)
env.close()
class MyActor1(MyActor):
def getPolicy(self, actor_index, actor_num):
import gym
from keras.optimizers import Adam
import traceback
import os
import sys
sys.path.append(os.path.join(os.path.dirname(__file__), '../'))
from src.common import seed_everything
seed_everything(43)
from src.r2d3 import Actor
from src.processor import AcrobotProcessor
from src.policy import EpsilonGreedy, AnnealingEpsilonGreedy
from src.image_model import DQNImageModel
from src.memory import PERRankBaseMemory, PERProportionalMemory
from src.common import InputType, LstmType, DuelingNetwork, LoggerType
from Lib import run_gym_rainbow, run_gym_r2d3, run_play, run_replay
ENV_NAME = "Acrobot-v1"
episode_save_dir = "tmp_{}.".format(ENV_NAME)
def create_parameter():
env = gym.make(ENV_NAME)
# ゲーム情報
print("action_space : " + str(env.action_space))