-
Notifications
You must be signed in to change notification settings - Fork 0
/
run_mujoco_sac.py
131 lines (101 loc) · 4.46 KB
/
run_mujoco_sac.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
import os
from absl import app, flags
from functools import partial
import numpy as np
import jax
import tqdm
import gym
import sac as learner
from jaxrl_m.wandb import setup_wandb, default_wandb_config, get_flag_dict
import wandb
from jaxrl_m.evaluation import supply_rng, evaluate, flatten, EpisodeMonitor
from jaxrl_m.dataset import ReplayBuffer
from ml_collections import config_flags
import pickle
from flax.training import checkpoints
FLAGS = flags.FLAGS
flags.DEFINE_string('env_name', 'HalfCheetah-v2', 'Environment name.')
flags.DEFINE_string('save_dir', None, 'Logging dir.')
flags.DEFINE_integer('seed', np.random.choice(1000000), 'Random seed.')
flags.DEFINE_integer('eval_episodes', 10,
'Number of episodes used for evaluation.')
flags.DEFINE_integer('log_interval', 1000, 'Logging interval.')
flags.DEFINE_integer('eval_interval', 10000, 'Eval interval.')
flags.DEFINE_integer('save_interval', 25000, 'Eval interval.')
flags.DEFINE_integer('batch_size', 256, 'Mini batch size.')
flags.DEFINE_integer('max_steps', int(1e6), 'Number of training steps.')
flags.DEFINE_integer('start_steps', int(1e4), 'Number of initial exploration steps.')
wandb_config = default_wandb_config()
wandb_config.update({
'project': 'd4rl_test',
'group': 'sac_test',
'name': 'sac_{env_name}',
})
config_flags.DEFINE_config_dict('wandb', wandb_config, lock_config=False)
config_flags.DEFINE_config_dict('config', learner.get_default_config(), lock_config=False)
def main(_):
# Create wandb logger
setup_wandb(FLAGS.config.to_dict(), **FLAGS.wandb)
if FLAGS.save_dir is not None:
FLAGS.save_dir = os.path.join(FLAGS.save_dir, wandb.run.project, wandb.config.exp_prefix, wandb.config.experiment_id)
os.makedirs(FLAGS.save_dir, exist_ok=True)
print(f'Saving config to {FLAGS.save_dir}/config.pkl')
with open(os.path.join(FLAGS.save_dir, 'config.pkl'), 'wb') as f:
pickle.dump(get_flag_dict(), f)
env = EpisodeMonitor(gym.make(FLAGS.env_name))
eval_env = EpisodeMonitor(gym.make(FLAGS.env_name))
example_transition = dict(
observations=env.observation_space.sample(),
actions=env.action_space.sample(),
rewards=0.0,
masks=1.0,
next_observations=env.observation_space.sample(),
)
replay_buffer = ReplayBuffer.create(example_transition, size=int(1e6))
agent = learner.create_learner(FLAGS.seed,
example_transition['observations'][None],
example_transition['actions'][None],
max_steps=FLAGS.max_steps,
**FLAGS.config)
exploration_metrics = dict()
obs = env.reset()
exploration_rng = jax.random.PRNGKey(0)
for i in tqdm.tqdm(range(1, FLAGS.max_steps + 1),
smoothing=0.1,
dynamic_ncols=True):
if i < FLAGS.start_steps:
action = env.action_space.sample()
else:
exploration_rng, key = jax.random.split(exploration_rng)
action = agent.sample_actions(obs, seed=key)
next_obs, reward, done, info = env.step(action)
mask = float(not done or 'TimeLimit.truncated' in info)
replay_buffer.add_transition(dict(
observations=obs,
actions=action,
rewards=reward,
masks=mask,
next_observations=next_obs,
))
obs = next_obs
if done:
exploration_metrics = {f'exploration/{k}': v for k, v in flatten(info).items()}
obs = env.reset()
if replay_buffer.size < FLAGS.start_steps:
continue
batch = replay_buffer.sample(FLAGS.batch_size)
agent, update_info = agent.update(batch)
if i % FLAGS.log_interval == 0:
train_metrics = {f'training/{k}': v for k, v in update_info.items()}
wandb.log(train_metrics, step=i)
wandb.log(exploration_metrics, step=i)
exploration_metrics = dict()
if i % FLAGS.eval_interval == 0:
policy_fn = partial(supply_rng(agent.sample_actions), temperature=0.0)
eval_info = evaluate(policy_fn, eval_env, num_episodes=FLAGS.eval_episodes)
eval_metrics = {f'evaluation/{k}': v for k, v in eval_info.items()}
wandb.log(eval_metrics, step=i)
if i % FLAGS.save_interval == 0 and FLAGS.save_dir is not None:
checkpoints.save_checkpoint(FLAGS.save_dir, agent, i)
if __name__ == '__main__':
app.run(main)