def get_experiment_environment(**args): from utils import setup_mpi_gpus, setup_tensorflow_session from baselines.common import set_global_seeds from gym.utils.seeding import hash_seed process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank() process_seed = hash_seed(process_seed, max_bytes=4) set_global_seeds(process_seed) setup_mpi_gpus() tf_context = setup_tensorflow_session() return tf_context
def get_experiment_environment(**args): from utils import setup_mpi_gpus, setup_tensorflow_session from baselines.common import set_global_seeds from gym.utils.seeding import hash_seed process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank() process_seed = hash_seed(process_seed, max_bytes=4) set_global_seeds(process_seed) setup_mpi_gpus() logger_context = logger.scoped_configure(dir=None, format_strs=['stdout', 'log', 'csv'] if MPI.COMM_WORLD.Get_rank() == 0 else ['log']) tf_context = setup_tensorflow_session() return logger_context, tf_context
def start_experiment(**args): from utils import setup_mpi_gpus setup_mpi_gpus() make_env = partial(make_env_all_params, add_monitor=True, args=args) trainer = Trainer(make_env=make_env, num_timesteps=args['num_timesteps'], hps=args, envs_per_process=args['envs_per_process']) log, tf_sess = get_experiment_environment(**args) with log, tf_sess: logdir = logger.get_dir() print("results will be saved to ", logdir) trainer.train()
def get_experiment_environment(**args): process_seed = 1234 + 1000 * MPI.COMM_WORLD.Get_rank() process_seed = hash_seed(process_seed, max_bytes=4) set_global_seeds(1234) setup_mpi_gpus() logger_context = logger.scoped_configure( dir='C:/Users/Elias/Desktop/savedunc/' + MODE + '_' + datetime.now().strftime('%Y_%m_%d_%H_%M_%S'), format_strs=['stdout', 'log', 'csv', 'tensorboard'] if MPI.COMM_WORLD.Get_rank() == 0 else ['log']) tf_context = setup_tensorflow_session() return logger_context, tf_context
def get_experiment_environment(**args): from utils import setup_mpi_gpus, setup_tensorflow_session from baselines.common import set_global_seeds from gym.utils.seeding import hash_seed process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank() process_seed = hash_seed(process_seed, max_bytes=4) set_global_seeds(process_seed) setup_mpi_gpus() logdir = './' + args["logdir"] + '/' + datetime.datetime.now().strftime( args["expID"] + "-openai-%Y-%m-%d-%H-%M-%S-%f") logger_context = logger.scoped_configure( dir=logdir, format_strs=['stdout', 'log', 'csv', 'tensorboard'] if MPI.COMM_WORLD.Get_rank() == 0 else ['log']) tf_context = setup_tensorflow_session() return logger_context, tf_context, logdir
def start_score(**args): from utils import setup_mpi_gpus setup_mpi_gpus() make_env = partial(make_env_all_params, add_monitor=True, args=args, sleep_multiple=0) scorer = Scorer(make_env=make_env, num_timesteps=args['num_timesteps'], hps=args, envs_per_process=args['envs_per_process']) log, tf_sess = get_experiment_environment(**args) with log, tf_sess: logdir = logger.get_dir() scorer.score()
def get_experiment_environment(**args): from utils import setup_mpi_gpus, setup_tensorflow_session from baselines.common import set_global_seeds from gym.utils.seeding import hash_seed process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank() process_seed = hash_seed(process_seed, max_bytes=4) set_global_seeds(process_seed) setup_mpi_gpus() time = datetime.datetime.now().strftime("%m-%d-%H-%M-%S") path_with_args = './logs/' + '_'.join([ time, args['exp_name'], args['env_kind'], args['feature_space'], str(args['envs_per_process']), str(args['train_discriminator']), str(args['discriminator_weighted']) ]) format_strs = ['stdout', 'log', 'csv', 'tensorboard' ] if MPI.COMM_WORLD.Get_rank() == 0 else ['log'] logger_context = logger.scoped_configure(dir=path_with_args, format_strs=format_strs) tf_context = setup_tensorflow_session() return logger_context, tf_context
def get_experiment_environment(**args): # 初始化 MPI 相关的量 from utils import setup_mpi_gpus, setup_tensorflow_session from baselines.common import set_global_seeds from gym.utils.seeding import hash_seed process_seed = args["seed"] + 1000 * MPI.COMM_WORLD.Get_rank() process_seed = hash_seed(process_seed, max_bytes=4) set_global_seeds(process_seed) setup_mpi_gpus() logger_dir = './logs/' + datetime.datetime.now().strftime( args["env"] + "-" + args["reward_type"] + "-" + str(args["nepochs_dvae"]) + "-" + str(args["stickyAtari"]) + "-%Y-%m-%d-%H-%M-%S-%f") logger_context = logger.scoped_configure( dir=logger_dir, format_strs=['stdout', 'log', 'csv'] if MPI.COMM_WORLD.Get_rank() == 0 else ['log']) tf_context = setup_tensorflow_session() # bai. 新增 saver 用于保存权重 saver = tf.train.Saver() return logger_context, tf_context, saver, logger_dir
#env.render() from utils import * from collections import deque import gym import cv2 import os import coinrun.main_utils as utils from coinrun import setup_utils, policies, wrappers, ppo2 from coinrun.config import Config #from gym.envs.classic_control import rendering from collections import deque import random from image_bco import ImageBCO utils.setup_mpi_gpus() setup_utils.setup_and_load() game = utils.make_general_env(1) game = wrappers.add_final_wrappers(game) game.reset() args.checkpoint = 'coin_ilpo' args.input_dir = 'final_models/coin' args.exp_dir = 'results/final_coin_bco' args.n_actions = 4 args.real_actions = 4 args.policy_lr = .0001 args.batch_size = 100 args.ngf = 15 states = [] next_states = []