def test_log(self):
log_dir = "./newtest/test_log.txt"
if os.path.exists(log_dir):
shutil.rmtree(log_dir)
logger = Logger(log_dir)
logger.log("test text")
logger.log_performance(1, 1)
logger.log_performance(2, 2)
logger.log_performance(3, 3)
logger.close_files()
logger.plot('aaa')
env.game.num_players, env.game.num_cards, episode_num))
# logger.log(f'\nTrain Agents:{get_agent_str(env_agent_list)}')
# logger.log(f'\nEval Agents:{get_agent_str(eval_agent_list)}')
for episode in range(episode_num):
# Generate data from the environment
trajectories, _ = env.run(is_training=True)
# Feed transitions into agent memory, and train the agent
for ts in trajectories[0]:
agent.feed(ts)
# Evaluate the performance. Play with random agents.
if episode % evaluate_every == 0:
logger.log_performance(env.timestep,
tournament(eval_env, evaluate_num)[0],
episode=episode)
# Save model
save_dir = 'models/mocsar_dqn_ra_pytorch'
if not os.path.exists(save_dir):
os.makedirs(save_dir)
state_dict = agent.get_state_dict()
logger.log('\n########## Pytorch Save model ##########')
logger.log('\n' + str(state_dict.keys()))
torch.save(state_dict, os.path.join(save_dir, 'model.pth'))
# Close files in the logger
logger.close_files()
# Plot the learning curve
state = env.reset()
for timestep in range(timesteps):
action = agent.step(state)
next_state, reward, done = env.step(action)
ts = (state, action, reward, next_state, done)
agent.feed(ts)
if timestep % evaluate_every == 0:
rewards = []
state = eval_env.reset()
for _ in range(evaluate_num):
action, _ = agent.eval_step(state)
_, reward, done = env.step(action)
if done:
rewards.append(reward)
logger.log_performance(env.timestep, np.mean(rewards))
# Close files in the logger
logger.close_files()
# Plot the learning curve
logger.plot('DQN')
# Save model
save_dir = 'models/uno_single_dqn'
if not os.path.exists(save_dir):
os.makedirs(save_dir)
saver = tf.train.Saver()
saver.save(sess, os.path.join(save_dir, 'model'))
