File size: 16,092 Bytes
9f589d7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
# docs and experiment results can be found at https://docs.cleanrl.dev/rl-algorithms/dqn/#dqn_ataripy
import argparse
import os
import random
import time
from distutils.util import strtobool

import gym
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from stable_baselines3.common.atari_wrappers import (
    ClipRewardEnv,
    EpisodicLifeEnv,
    FireResetEnv,
    MaxAndSkipEnv,
    NoopResetEnv,
)
from stable_baselines3.common.buffers import ReplayBuffer
from torch.utils.tensorboard import SummaryWriter


def parse_args():
    # fmt: off
    parser = argparse.ArgumentParser()
    parser.add_argument("--exp-name", type=str, default=os.path.basename(__file__).rstrip(".py"),
        help="the name of this experiment")
    parser.add_argument("--seed", type=int, default=1,
        help="seed of the experiment")
    parser.add_argument("--torch-deterministic", type=lambda x: bool(strtobool(x)), default=True, nargs="?", const=True,
        help="if toggled, `torch.backends.cudnn.deterministic=False`")
    parser.add_argument("--cuda", type=lambda x: bool(strtobool(x)), default=True, nargs="?", const=True,
        help="if toggled, cuda will be enabled by default")
    parser.add_argument("--track", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
        help="if toggled, this experiment will be tracked with Weights and Biases")
    parser.add_argument("--wandb-project-name", type=str, default="cleanRL",
        help="the wandb's project name")
    parser.add_argument("--wandb-entity", type=str, default=None,
        help="the entity (team) of wandb's project")
    parser.add_argument("--capture-video", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
        help="whether to capture videos of the agent performances (check out `videos` folder)")
    parser.add_argument("--save-model", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
        help="whether to save model into the `runs/{run_name}` folder")
    parser.add_argument("--upload-model", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
        help="whether to upload the saved model to huggingface")
    parser.add_argument("--hf-entity", type=str, default="",
        help="the user or org name of the model repository from the Hugging Face Hub")

    # Algorithm specific arguments
    parser.add_argument("--env-id", type=str, default="PongNoFrameskip-v4",
        help="the id of the environment")
    parser.add_argument("--total-timesteps", type=int, default=5000000,
        help="total timesteps of the experiments")
    parser.add_argument("--learning-rate", type=float, default=0.0001,
        help="the learning rate of the optimizer")
    parser.add_argument("--max-gradient-norm", type=float, default=float('inf'),
        help="gradient clipping value")
    parser.add_argument("--double-learning", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
        help="enable double learning DDQN")
    parser.add_argument("--buffer-size", type=int, default=1000000,
        help="the replay memory buffer size")
    parser.add_argument("--gamma", type=float, default=0.99,
        help="the discount factor gamma")
    parser.add_argument("--target-tau", type=float, default=1.0,
        help="the target network update rate")
    parser.add_argument("--policy-tau", type=float, default=1.0,
        help="the target network update rate")
    parser.add_argument("--target-network-frequency", type=int, default=1000,
        help="the timesteps it takes to update the target network")
    parser.add_argument("--policy-network-frequency", type=int, default=10000,
        help="the timesteps it takes to update the policy network")
    parser.add_argument("--batch-size", type=int, default=32,
        help="the batch size of sample from the reply memory")
    parser.add_argument("--start-e", type=float, default=1.0,
        help="the starting epsilon for exploration")
    parser.add_argument("--end-e", type=float, default=0.01,
        help="the ending epsilon for exploration")
    parser.add_argument("--exploration-fraction", type=float, default=0.2,
        help="the fraction of `total-timesteps` it takes from start-e to go end-e")
    parser.add_argument("--learning-starts", type=int, default=10000,
        help="timestep to start learning")
    parser.add_argument("--train-frequency", type=int, default=1,
        help="the frequency of training")
    args = parser.parse_args()
    # fmt: on
    return args


def make_env(env_id, seed, idx, capture_video, run_name):
    def thunk():
        env = gym.make(env_id)
        env = gym.wrappers.RecordEpisodeStatistics(env)
        if capture_video:
            if idx == 0:
                env = gym.wrappers.RecordVideo(env, f"videos/{run_name}")
        env = NoopResetEnv(env, noop_max=30)
        env = MaxAndSkipEnv(env, skip=4)
        env = EpisodicLifeEnv(env)
        if "FIRE" in env.unwrapped.get_action_meanings():
            env = FireResetEnv(env)
        env = ClipRewardEnv(env)
        env = gym.wrappers.ResizeObservation(env, (84, 84))
        env = gym.wrappers.GrayScaleObservation(env)
        env = gym.wrappers.FrameStack(env, 4)
        env.seed(seed)
        env.action_space.seed(seed)
        env.observation_space.seed(seed)
        return env

    return thunk


# ALGO LOGIC: initialize agent here:
class QNetwork(nn.Module):
    def __init__(self, env):
        super().__init__()
        self.network = nn.Sequential(
            nn.Conv2d(4, 32, 8, stride=4),
            nn.ReLU(),
            nn.Conv2d(32, 64, 4, stride=2),
            nn.ReLU(),
            nn.Conv2d(64, 64, 3, stride=1),
            nn.ReLU(),
            nn.Flatten(),
            nn.Linear(3136, 512),
            nn.ReLU(),
            nn.Linear(512, env.single_action_space.n),
        )

    def forward(self, x):
        return self.network(x / 255.0)


def linear_schedule(start_e: float, end_e: float, duration: int, t: int):
    slope = (end_e - start_e) / duration
    return max(slope * t + start_e, end_e)


if __name__ == "__main__":
    args = parse_args()
    run_name = f"{args.env_id}__{args.exp_name}__{args.seed}__{int(time.time())}"
    if args.track:
        import wandb

        args.alg_type = os.path.basename(__file__)
        wandb_sess = wandb.init(
            project=args.wandb_project_name,
            entity=args.wandb_entity,
            config=vars(args),
            save_code=True,
            # group='string',
            name=run_name,
            sync_tensorboard=False,
            monitor_gym=True,
        )
    writer = SummaryWriter(f"runs/{run_name}")
    writer.add_text(
        "hyperparameters",
        "|param|value|\n|-|-|\n%s" % ("\n".join([f"|{key}|{value}|" for key, value in vars(args).items()])),
    )

    def log_value(name: str, x: float, y: int):
        # writer.add_scalar(name, x, y)
        wandb.log({name: x, "global_step": y})

    # TRY NOT TO MODIFY: seeding
    random.seed(args.seed)
    np.random.seed(args.seed)
    torch.manual_seed(args.seed)
    torch.backends.cudnn.deterministic = args.torch_deterministic

    device = torch.device("cuda" if torch.cuda.is_available() and args.cuda else "cpu")

    # env setup
    envs = gym.vector.SyncVectorEnv([make_env(args.env_id, args.seed, 0, args.capture_video, run_name)])
    assert isinstance(envs.single_action_space, gym.spaces.Discrete), "only discrete action space is supported"

    q_network = QNetwork(envs).to(device)
    optimizer = optim.RMSprop(q_network.parameters(), lr=args.learning_rate)
    target_network = QNetwork(envs).to(device)
    policy_network = QNetwork(envs).to(device)
    target_network.load_state_dict(q_network.state_dict())
    policy_network.load_state_dict(q_network.state_dict())

    rb = ReplayBuffer(
        args.buffer_size,
        envs.single_observation_space,
        envs.single_action_space,
        device,
        optimize_memory_usage=True,
        handle_timeout_termination=True,
    )
    start_time = time.time()
    target_update_counter = 0
    policy_update_counter = 0
    episode_returns = []

    # TRY NOT TO MODIFY: start the game
    obs = envs.reset()
    for global_step in range(args.total_timesteps):
        # ALGO LOGIC: put action logic here
        epsilon = linear_schedule(args.start_e, args.end_e, args.exploration_fraction * args.total_timesteps, global_step)

        if random.random() < epsilon:
            actions = np.array([envs.single_action_space.sample() for _ in range(envs.num_envs)])
        else:
            q_values = policy_network(torch.Tensor(obs).to(device))
            actions = torch.argmax(q_values, dim=1).cpu().numpy()

        # TRY NOT TO MODIFY: execute the game and log data.
        next_obs, rewards, dones, infos = envs.step(actions)

        # TRY NOT TO MODIFY: record rewards for plotting purposes
        for info in infos:
            if "episode" in info.keys():
                episode_returns.append(info['episode']['r'])
                episode_returns = episode_returns[-100:]
                print(f"step={global_step}, return={info['episode']['r']}, sps={int(global_step / (time.time() - start_time))}")
                log_value("perf/episodic_return", info["episode"]["r"], global_step)
                log_value("perf/episodic_return_mean_100", np.mean(episode_returns), global_step)
                log_value("perf/episodic_return_std_100", np.std(episode_returns), global_step)
                log_value("debug/episodic_length", info["episode"]["l"], global_step)
                log_value("ex2/epsilon", epsilon, global_step)
                break

        # TRY NOT TO MODIFY: save data to reply buffer; handle `terminal_observation`
        real_next_obs = next_obs.copy()
        for idx, d in enumerate(dones):
            if d:
                real_next_obs[idx] = infos[idx]["terminal_observation"]
        rb.add(obs, real_next_obs, actions, rewards, dones, infos)

        # TRY NOT TO MODIFY: CRUCIAL step easy to overlook
        obs = next_obs

        # ALGO LOGIC: training.
        if global_step > args.learning_starts:
            # NOTE: Current code does not work with train_frequency != 1
            if global_step % args.train_frequency == 0:
                data = rb.sample(args.batch_size)
                with torch.no_grad():
                    if args.double_learning:
                        argmax_a = q_network(data.next_observations).max(1)[1].unsqueeze(1)
                    else:
                        argmax_a = target_network(data.next_observations).max(1)[1].unsqueeze(1)

                    target_max = target_network(data.next_observations).gather(1, argmax_a).squeeze()
                    td_target = data.rewards.flatten() + args.gamma * target_max * (1 - data.dones.flatten())

                old_val = q_network(data.observations).gather(1, data.actions).squeeze()
                loss = F.mse_loss(td_target, old_val)

                if global_step % 100 == 0:

                    prev = old_val.detach().cpu().numpy()
                    new = td_target.detach().cpu().numpy()
                    diff, a_diff = new-prev, np.abs(new-prev)

                    mean, a_mean = np.mean(diff), np.mean(a_diff)
                    median, a_median = np.median(diff), np.median(a_diff)
                    maximum, a_maximum = np.max(diff), np.max(a_diff)
                    minimum, a_minimum = np.min(diff), np.min(a_diff)
                    std, a_std = np.std(diff), np.std(a_diff)
                    below, a_below = mean - std, a_mean - a_std
                    above, a_above = mean + std, a_mean + a_std
                    pu_scalar, a_pu_scalar = 2 * mean / maximum, 2 * a_mean / a_maximum
                    policy_frequency_scalar_ratio = args.policy_network_frequency * pu_scalar
                    a_policy_frequency_scalar_ratio = args.policy_network_frequency * a_pu_scalar

                    log_value("losses/td_loss", loss, global_step)
                    log_value("losses/q_values", old_val.mean().item(), global_step)
                    log_value("td/mean", mean, global_step)
                    log_value("td/a_mean", a_mean, global_step)
                    log_value("td/median", median, global_step)
                    log_value("td/a_median", a_median, global_step)
                    log_value("td/max", maximum, global_step)
                    log_value("td/a_max", a_maximum, global_step)
                    log_value("td/min", minimum, global_step)
                    log_value("td/a_min", a_minimum, global_step)
                    log_value("td/std", std, global_step)
                    log_value("td/a_std", a_std, global_step)
                    log_value("td/below", below, global_step)
                    log_value("td/a_below", a_below, global_step)
                    log_value("td/above", above, global_step)
                    log_value("td/a_above", a_above, global_step)
                    log_value("alg/pu_scalar", pu_scalar, global_step)
                    log_value("alg/a_pu_scalar", a_pu_scalar, global_step)
                    log_value("alg/policy_frequency_scalar_ratio", policy_frequency_scalar_ratio, global_step)
                    log_value("alg/a_policy_frequency_scalar_ratio", a_policy_frequency_scalar_ratio, global_step)
                    log_value("debug/steps_per_second", int(global_step / (time.time() - start_time)), global_step)

                # optimize the model
                optimizer.zero_grad()
                loss.backward()
                torch.nn.utils.clip_grad_norm_(q_network.parameters(),
                                               args.max_gradient_norm)
                optimizer.step()

            # update target network
            if global_step % args.target_network_frequency == 0:
                target_update_counter += 1
                for target_network_param, q_network_param in zip(target_network.parameters(), q_network.parameters()):
                    target_network_param.data.copy_(
                        args.target_tau * q_network_param.data + (1.0 - args.target_tau) * target_network_param.data
                    )

            # update policy network
            if global_step % args.policy_network_frequency == 0:
                policy_update_counter += 1
                for policy_network_param, q_network_param in zip(policy_network.parameters(), q_network.parameters()):
                    policy_network_param.data.copy_(
                        args.policy_tau * q_network_param.data + (1.0 - args.policy_tau) * policy_network_param.data
                    )

            if global_step % 100 == 0:
                log_value("alg/n_target_update", target_update_counter, global_step)
                log_value("alg/n_policy_update", policy_update_counter, global_step)

    if args.save_model:
        model_path = f"runs/{run_name}/{args.exp_name}.cleanrl_model"
        torch.save(policy_network.state_dict(), model_path)
        print(f"model saved to {model_path}")
        from cleanrl_utils.evals.dqn_eval import evaluate

        episodic_returns = evaluate(
            model_path,
            make_env,
            args.env_id,
            eval_episodes=10,
            run_name=f"{run_name}-eval",
            Model=QNetwork,
            device=device,
            epsilon=0.05,
        )
        for idx, episodic_return in enumerate(episodic_returns):
            log_value("eval/episodic_return", episodic_return, idx)


        if args.upload_model:
            from cleanrl_utils.huggingface import push_to_hub

            repo_name = f"{args.env_id}-{args.exp_name}-seed{args.seed}"
            repo_id = f"{args.hf_entity}/{repo_name}" if args.hf_entity else repo_name
            push_to_hub(args, np.mean(episode_returns), repo_id, "DQPN_freq", f"runs/{run_name}", f"videos/{run_name}-eval")

    wandb_sess.finish()
    envs.close()
    writer.close()