def load(path, num_cpu=16):
with open(path, "rb") as f:
model_data, act_params = dill.load(f)
act = distdeepq.build_act(**act_params)
sess = U.make_session(num_cpu=num_cpu)
sess.__enter__()
with tempfile.TemporaryDirectory() as td:
arc_path = os.path.join(td, "packed.zip")
with open(arc_path, "wb") as f:
f.write(model_data)
zipfile.ZipFile(arc_path, 'r', zipfile.ZIP_DEFLATED).extractall(td)
U.load_state(os.path.join(td, "model"))
return ActWrapper(act, act_params)
def main():
set_global_seeds(1)
args = parse_args()
with U.make_session(4) as sess: # noqa
_, env = make_env(args.env)
model_parent_path = distdeepq.parent_path(args.model_dir)
old_args = json.load(open(model_parent_path + '/args.json'))
act = distdeepq.build_act(make_obs_ph=lambda name: U.Uint8Input(
env.observation_space.shape, name=name),
p_dist_func=distdeepq.models.atari_model(),
num_actions=env.action_space.n,
dist_params={
'Vmin': old_args['vmin'],
'Vmax': old_args['vmax'],
'nb_atoms': old_args['nb_atoms']
})
U.load_state(os.path.join(args.model_dir, "saved"))
wang2015_eval(args.env, act, stochastic=args.stochastic)
obs = env.reset()
if len(info["rewards"]) > num_episodes:
if len(info["rewards"]) == 1 and video_recorder.enabled:
# save video of first episode
print("Saved video.")
video_recorder.close()
video_recorder.enabled = False
print(info["rewards"][-1])
num_episodes = len(info["rewards"])
if __name__ == '__main__':
with U.make_session(4) as sess:
args = parse_args()
env = make_env(args.env)
model_parent_path = distdeepq.parent_path(args.model_dir)
old_args = json.load(open(model_parent_path + '/args.json'))
act = distdeepq.build_act(make_obs_ph=lambda name: U.Uint8Input(
env.observation_space.shape, name=name),
p_dist_func=distdeepq.models.atari_model(),
num_actions=env.action_space.n,
dist_params={
'Vmin': old_args['vmin'],
'Vmax': old_args['vmax'],
'nb_atoms': old_args['nb_atoms']
})
U.load_state(os.path.join(args.model_dir, "saved"))
play(env, act, args.stochastic, args.video)
if len(info["rewards"]) > num_episodes:
if len(info["rewards"]) == 1 and video_recorder.enabled:
# save video of first episode
print("Saved video.")
video_recorder.close()
video_recorder.enabled = False
print(info["rewards"][-1])
num_episodes = len(info["rewards"])
if __name__ == '__main__':
with U.make_session(4) as sess:
args = parse_args()
env, _ = distdeepq.make_env(args.env)
model_parent_path = distdeepq.parent_path(args.model_dir)
old_args = json.load(open(model_parent_path + '/args.json'))
dist_params = {
'Vmin': old_args['vmin'],
'Vmax': old_args['vmax'],
'nb_atoms': old_args['nb_atoms']
}
act = distdeepq.build_act(make_obs_ph=lambda name: U.Uint8Input(
env.observation_space.shape, name=name),
p_dist_func=distdeepq.models.atari_model(),
num_actions=env.action_space.n,
dist_params=dist_params)
U.load_state(os.path.join(args.model_dir, "saved"))
play(env, act, args.stochastic, args.video)