179 lines
6.4 KiB
Python
Executable File
179 lines
6.4 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
|
|
import argparse
|
|
import datetime
|
|
import os
|
|
import pprint
|
|
|
|
import numpy as np
|
|
import torch
|
|
from mujoco_env import make_mujoco_env
|
|
from torch.utils.tensorboard import SummaryWriter
|
|
|
|
from tianshou.data import Collector, ReplayBuffer, VectorReplayBuffer
|
|
from tianshou.exploration import GaussianNoise
|
|
from tianshou.policy import DDPGPolicy
|
|
from tianshou.trainer import OffpolicyTrainer
|
|
from tianshou.utils import TensorboardLogger, WandbLogger
|
|
from tianshou.utils.net.common import Net
|
|
from tianshou.utils.net.continuous import Actor, Critic
|
|
|
|
|
|
def get_args():
|
|
parser = argparse.ArgumentParser()
|
|
parser.add_argument("--task", type=str, default="Ant-v4")
|
|
parser.add_argument("--seed", type=int, default=0)
|
|
parser.add_argument("--buffer-size", type=int, default=1000000)
|
|
parser.add_argument("--hidden-sizes", type=int, nargs="*", default=[256, 256])
|
|
parser.add_argument("--actor-lr", type=float, default=1e-3)
|
|
parser.add_argument("--critic-lr", type=float, default=1e-3)
|
|
parser.add_argument("--gamma", type=float, default=0.99)
|
|
parser.add_argument("--tau", type=float, default=0.005)
|
|
parser.add_argument("--exploration-noise", type=float, default=0.1)
|
|
parser.add_argument("--start-timesteps", type=int, default=25000)
|
|
parser.add_argument("--epoch", type=int, default=200)
|
|
parser.add_argument("--step-per-epoch", type=int, default=5000)
|
|
parser.add_argument("--step-per-collect", type=int, default=1)
|
|
parser.add_argument("--update-per-step", type=int, default=1)
|
|
parser.add_argument("--n-step", type=int, default=1)
|
|
parser.add_argument("--batch-size", type=int, default=256)
|
|
parser.add_argument("--training-num", type=int, default=1)
|
|
parser.add_argument("--test-num", type=int, default=10)
|
|
parser.add_argument("--logdir", type=str, default="log")
|
|
parser.add_argument("--render", type=float, default=0.0)
|
|
parser.add_argument(
|
|
"--device",
|
|
type=str,
|
|
default="cuda" if torch.cuda.is_available() else "cpu",
|
|
)
|
|
parser.add_argument("--resume-path", type=str, default=None)
|
|
parser.add_argument("--resume-id", type=str, default=None)
|
|
parser.add_argument(
|
|
"--logger",
|
|
type=str,
|
|
default="tensorboard",
|
|
choices=["tensorboard", "wandb"],
|
|
)
|
|
parser.add_argument("--wandb-project", type=str, default="mujoco.benchmark")
|
|
parser.add_argument(
|
|
"--watch",
|
|
default=False,
|
|
action="store_true",
|
|
help="watch the play of pre-trained policy only",
|
|
)
|
|
return parser.parse_args()
|
|
|
|
|
|
def test_ddpg(args=get_args()):
|
|
env, train_envs, test_envs = make_mujoco_env(
|
|
args.task,
|
|
args.seed,
|
|
args.training_num,
|
|
args.test_num,
|
|
obs_norm=False,
|
|
)
|
|
args.state_shape = env.observation_space.shape or env.observation_space.n
|
|
args.action_shape = env.action_space.shape or env.action_space.n
|
|
args.max_action = env.action_space.high[0]
|
|
args.exploration_noise = args.exploration_noise * args.max_action
|
|
print("Observations shape:", args.state_shape)
|
|
print("Actions shape:", args.action_shape)
|
|
print("Action range:", np.min(env.action_space.low), np.max(env.action_space.high))
|
|
# seed
|
|
np.random.seed(args.seed)
|
|
torch.manual_seed(args.seed)
|
|
# model
|
|
net_a = Net(args.state_shape, hidden_sizes=args.hidden_sizes, device=args.device)
|
|
actor = Actor(net_a, args.action_shape, max_action=args.max_action, device=args.device).to(
|
|
args.device,
|
|
)
|
|
actor_optim = torch.optim.Adam(actor.parameters(), lr=args.actor_lr)
|
|
net_c = Net(
|
|
args.state_shape,
|
|
args.action_shape,
|
|
hidden_sizes=args.hidden_sizes,
|
|
concat=True,
|
|
device=args.device,
|
|
)
|
|
critic = Critic(net_c, device=args.device).to(args.device)
|
|
critic_optim = torch.optim.Adam(critic.parameters(), lr=args.critic_lr)
|
|
policy = DDPGPolicy(
|
|
actor=actor,
|
|
actor_optim=actor_optim,
|
|
critic=critic,
|
|
critic_optim=critic_optim,
|
|
tau=args.tau,
|
|
gamma=args.gamma,
|
|
exploration_noise=GaussianNoise(sigma=args.exploration_noise),
|
|
estimation_step=args.n_step,
|
|
action_space=env.action_space,
|
|
)
|
|
|
|
# load a previous policy
|
|
if args.resume_path:
|
|
policy.load_state_dict(torch.load(args.resume_path, map_location=args.device))
|
|
print("Loaded agent from: ", args.resume_path)
|
|
|
|
# collector
|
|
if args.training_num > 1:
|
|
buffer = VectorReplayBuffer(args.buffer_size, len(train_envs))
|
|
else:
|
|
buffer = ReplayBuffer(args.buffer_size)
|
|
train_collector = Collector(policy, train_envs, buffer, exploration_noise=True)
|
|
test_collector = Collector(policy, test_envs)
|
|
train_collector.collect(n_step=args.start_timesteps, random=True)
|
|
|
|
# log
|
|
now = datetime.datetime.now().strftime("%y%m%d-%H%M%S")
|
|
args.algo_name = "ddpg"
|
|
log_name = os.path.join(args.task, args.algo_name, str(args.seed), now)
|
|
log_path = os.path.join(args.logdir, log_name)
|
|
|
|
# logger
|
|
if args.logger == "wandb":
|
|
logger = WandbLogger(
|
|
save_interval=1,
|
|
name=log_name.replace(os.path.sep, "__"),
|
|
run_id=args.resume_id,
|
|
config=args,
|
|
project=args.wandb_project,
|
|
)
|
|
writer = SummaryWriter(log_path)
|
|
writer.add_text("args", str(args))
|
|
if args.logger == "tensorboard":
|
|
logger = TensorboardLogger(writer)
|
|
else: # wandb
|
|
logger.load(writer)
|
|
|
|
def save_best_fn(policy):
|
|
torch.save(policy.state_dict(), os.path.join(log_path, "policy.pth"))
|
|
|
|
if not args.watch:
|
|
# trainer
|
|
result = OffpolicyTrainer(
|
|
policy=policy,
|
|
train_collector=train_collector,
|
|
test_collector=test_collector,
|
|
max_epoch=args.epoch,
|
|
step_per_epoch=args.step_per_epoch,
|
|
step_per_collect=args.step_per_collect,
|
|
episode_per_test=args.test_num,
|
|
batch_size=args.batch_size,
|
|
save_best_fn=save_best_fn,
|
|
logger=logger,
|
|
update_per_step=args.update_per_step,
|
|
test_in_train=False,
|
|
).run()
|
|
pprint.pprint(result)
|
|
|
|
# Let's watch its performance!
|
|
policy.eval()
|
|
test_envs.seed(args.seed)
|
|
test_collector.reset()
|
|
result = test_collector.collect(n_episode=args.test_num, render=args.render)
|
|
print(f"Final reward: {result.returns_stat.mean}, length: {result.lens_stat.mean}")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
test_ddpg()
|