File size: 12,487 Bytes
88739bd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
352
353
354
import numpy as np
import torch
import torch.nn as nn

from dataclasses import asdict, dataclass, field
from time import perf_counter
from torch.optim import Adam
from torch.utils.tensorboard.writer import SummaryWriter
from typing import List, Optional, NamedTuple, TypeVar

from rl_algo_impls.shared.algorithm import Algorithm
from rl_algo_impls.shared.callbacks.callback import Callback
from rl_algo_impls.shared.gae import compute_advantage, compute_rtg_and_advantage
from rl_algo_impls.shared.policy.on_policy import ActorCritic
from rl_algo_impls.shared.schedule import (
    constant_schedule,
    linear_schedule,
    update_learning_rate,
)
from rl_algo_impls.shared.trajectory import Trajectory, TrajectoryAccumulator
from rl_algo_impls.wrappers.vectorable_wrapper import VecEnv, VecEnvObs


@dataclass
class PPOTrajectory(Trajectory):
    logp_a: List[float] = field(default_factory=list)

    def add(
        self,
        obs: np.ndarray,
        act: np.ndarray,
        next_obs: np.ndarray,
        rew: float,
        terminated: bool,
        v: float,
        logp_a: float,
    ):
        super().add(obs, act, next_obs, rew, terminated, v)
        self.logp_a.append(logp_a)


class PPOTrajectoryAccumulator(TrajectoryAccumulator):
    def __init__(self, num_envs: int) -> None:
        super().__init__(num_envs, PPOTrajectory)

    def step(
        self,
        obs: VecEnvObs,
        action: np.ndarray,
        next_obs: VecEnvObs,
        reward: np.ndarray,
        done: np.ndarray,
        val: np.ndarray,
        logp_a: np.ndarray,
    ) -> None:
        super().step(obs, action, next_obs, reward, done, val, logp_a)


class TrainStepStats(NamedTuple):
    loss: float
    pi_loss: float
    v_loss: float
    entropy_loss: float
    approx_kl: float
    clipped_frac: float
    val_clipped_frac: float


@dataclass
class TrainStats:
    loss: float
    pi_loss: float
    v_loss: float
    entropy_loss: float
    approx_kl: float
    clipped_frac: float
    val_clipped_frac: float
    explained_var: float

    def __init__(self, step_stats: List[TrainStepStats], explained_var: float) -> None:
        self.loss = np.mean([s.loss for s in step_stats]).item()
        self.pi_loss = np.mean([s.pi_loss for s in step_stats]).item()
        self.v_loss = np.mean([s.v_loss for s in step_stats]).item()
        self.entropy_loss = np.mean([s.entropy_loss for s in step_stats]).item()
        self.approx_kl = np.mean([s.approx_kl for s in step_stats]).item()
        self.clipped_frac = np.mean([s.clipped_frac for s in step_stats]).item()
        self.val_clipped_frac = np.mean([s.val_clipped_frac for s in step_stats]).item()
        self.explained_var = explained_var

    def write_to_tensorboard(self, tb_writer: SummaryWriter, global_step: int) -> None:
        for name, value in asdict(self).items():
            tb_writer.add_scalar(f"losses/{name}", value, global_step=global_step)

    def __repr__(self) -> str:
        return " | ".join(
            [
                f"Loss: {round(self.loss, 2)}",
                f"Pi L: {round(self.pi_loss, 2)}",
                f"V L: {round(self.v_loss, 2)}",
                f"E L: {round(self.entropy_loss, 2)}",
                f"Apx KL Div: {round(self.approx_kl, 2)}",
                f"Clip Frac: {round(self.clipped_frac, 2)}",
                f"Val Clip Frac: {round(self.val_clipped_frac, 2)}",
            ]
        )


PPOSelf = TypeVar("PPOSelf", bound="PPO")


class PPO(Algorithm):
    def __init__(
        self,
        policy: ActorCritic,
        env: VecEnv,
        device: torch.device,
        tb_writer: SummaryWriter,
        learning_rate: float = 3e-4,
        learning_rate_decay: str = "none",
        n_steps: int = 2048,
        batch_size: int = 64,
        n_epochs: int = 10,
        gamma: float = 0.99,
        gae_lambda: float = 0.95,
        clip_range: float = 0.2,
        clip_range_decay: str = "none",
        clip_range_vf: Optional[float] = None,
        clip_range_vf_decay: str = "none",
        normalize_advantage: bool = True,
        ent_coef: float = 0.0,
        ent_coef_decay: str = "none",
        vf_coef: float = 0.5,
        ppo2_vf_coef_halving: bool = False,
        max_grad_norm: float = 0.5,
        update_rtg_between_epochs: bool = False,
        sde_sample_freq: int = -1,
    ) -> None:
        super().__init__(policy, env, device, tb_writer)
        self.policy = policy

        self.gamma = gamma
        self.gae_lambda = gae_lambda
        self.optimizer = Adam(self.policy.parameters(), lr=learning_rate, eps=1e-7)
        self.lr_schedule = (
            linear_schedule(learning_rate, 0)
            if learning_rate_decay == "linear"
            else constant_schedule(learning_rate)
        )
        self.max_grad_norm = max_grad_norm
        self.clip_range_schedule = (
            linear_schedule(clip_range, 0)
            if clip_range_decay == "linear"
            else constant_schedule(clip_range)
        )
        self.clip_range_vf_schedule = None
        if clip_range_vf:
            self.clip_range_vf_schedule = (
                linear_schedule(clip_range_vf, 0)
                if clip_range_vf_decay == "linear"
                else constant_schedule(clip_range_vf)
            )
        self.normalize_advantage = normalize_advantage
        self.ent_coef_schedule = (
            linear_schedule(ent_coef, 0)
            if ent_coef_decay == "linear"
            else constant_schedule(ent_coef)
        )
        self.vf_coef = vf_coef
        self.ppo2_vf_coef_halving = ppo2_vf_coef_halving

        self.n_steps = n_steps
        self.batch_size = batch_size
        self.n_epochs = n_epochs
        self.sde_sample_freq = sde_sample_freq

        self.update_rtg_between_epochs = update_rtg_between_epochs

    def learn(
        self: PPOSelf,
        total_timesteps: int,
        callback: Optional[Callback] = None,
    ) -> PPOSelf:
        obs = self.env.reset()
        ts_elapsed = 0
        while ts_elapsed < total_timesteps:
            start_time = perf_counter()
            accumulator = self._collect_trajectories(obs)
            rollout_steps = self.n_steps * self.env.num_envs
            ts_elapsed += rollout_steps
            progress = ts_elapsed / total_timesteps
            train_stats = self.train(accumulator.all_trajectories, progress, ts_elapsed)
            train_stats.write_to_tensorboard(self.tb_writer, ts_elapsed)
            end_time = perf_counter()
            self.tb_writer.add_scalar(
                "train/steps_per_second",
                rollout_steps / (end_time - start_time),
                ts_elapsed,
            )
            if callback:
                callback.on_step(timesteps_elapsed=rollout_steps)

        return self

    def _collect_trajectories(self, obs: VecEnvObs) -> PPOTrajectoryAccumulator:
        self.policy.eval()
        accumulator = PPOTrajectoryAccumulator(self.env.num_envs)
        self.policy.reset_noise()
        for i in range(self.n_steps):
            if self.sde_sample_freq > 0 and i > 0 and i % self.sde_sample_freq == 0:
                self.policy.reset_noise()
            action, value, logp_a, clamped_action = self.policy.step(obs)
            next_obs, reward, done, _ = self.env.step(clamped_action)
            accumulator.step(obs, action, next_obs, reward, done, value, logp_a)
            obs = next_obs
        return accumulator

    def train(
        self, trajectories: List[PPOTrajectory], progress: float, timesteps_elapsed: int
    ) -> TrainStats:
        self.policy.train()
        learning_rate = self.lr_schedule(progress)
        update_learning_rate(self.optimizer, learning_rate)
        self.tb_writer.add_scalar(
            "charts/learning_rate",
            self.optimizer.param_groups[0]["lr"],
            timesteps_elapsed,
        )

        pi_clip = self.clip_range_schedule(progress)
        self.tb_writer.add_scalar("charts/pi_clip", pi_clip, timesteps_elapsed)
        if self.clip_range_vf_schedule:
            v_clip = self.clip_range_vf_schedule(progress)
            self.tb_writer.add_scalar("charts/v_clip", v_clip, timesteps_elapsed)
        else:
            v_clip = None
        ent_coef = self.ent_coef_schedule(progress)
        self.tb_writer.add_scalar("charts/ent_coef", ent_coef, timesteps_elapsed)

        obs = torch.as_tensor(
            np.concatenate([np.array(t.obs) for t in trajectories]), device=self.device
        )
        act = torch.as_tensor(
            np.concatenate([np.array(t.act) for t in trajectories]), device=self.device
        )
        rtg, adv = compute_rtg_and_advantage(
            trajectories, self.policy, self.gamma, self.gae_lambda, self.device
        )
        orig_v = torch.as_tensor(
            np.concatenate([np.array(t.v) for t in trajectories]), device=self.device
        )
        orig_logp_a = torch.as_tensor(
            np.concatenate([np.array(t.logp_a) for t in trajectories]),
            device=self.device,
        )

        step_stats = []
        for _ in range(self.n_epochs):
            step_stats.clear()
            if self.update_rtg_between_epochs:
                rtg, adv = compute_rtg_and_advantage(
                    trajectories, self.policy, self.gamma, self.gae_lambda, self.device
                )
            else:
                adv = compute_advantage(
                    trajectories, self.policy, self.gamma, self.gae_lambda, self.device
                )
            idxs = torch.randperm(len(obs))
            for i in range(0, len(obs), self.batch_size):
                mb_idxs = idxs[i : i + self.batch_size]
                mb_adv = adv[mb_idxs]
                if self.normalize_advantage:
                    mb_adv = (mb_adv - mb_adv.mean(-1)) / (mb_adv.std(-1) + 1e-8)
                self.policy.reset_noise(self.batch_size)
                step_stats.append(
                    self._train_step(
                        pi_clip,
                        v_clip,
                        ent_coef,
                        obs[mb_idxs],
                        act[mb_idxs],
                        rtg[mb_idxs],
                        mb_adv,
                        orig_v[mb_idxs],
                        orig_logp_a[mb_idxs],
                    )
                )

        y_pred, y_true = orig_v.cpu().numpy(), rtg.cpu().numpy()
        var_y = np.var(y_true).item()
        explained_var = (
            np.nan if var_y == 0 else 1 - np.var(y_true - y_pred).item() / var_y
        )

        return TrainStats(step_stats, explained_var)

    def _train_step(
        self,
        pi_clip: float,
        v_clip: Optional[float],
        ent_coef: float,
        obs: torch.Tensor,
        act: torch.Tensor,
        rtg: torch.Tensor,
        adv: torch.Tensor,
        orig_v: torch.Tensor,
        orig_logp_a: torch.Tensor,
    ) -> TrainStepStats:
        logp_a, entropy, v = self.policy(obs, act)
        logratio = logp_a - orig_logp_a
        ratio = torch.exp(logratio)
        clip_ratio = torch.clamp(ratio, min=1 - pi_clip, max=1 + pi_clip)
        pi_loss = torch.maximum(-ratio * adv, -clip_ratio * adv).mean()

        v_loss_unclipped = (v - rtg) ** 2
        if v_clip:
            v_loss_clipped = (
                orig_v + torch.clamp(v - orig_v, -v_clip, v_clip) - rtg
            ) ** 2
            v_loss = torch.max(v_loss_unclipped, v_loss_clipped).mean()
        else:
            v_loss = v_loss_unclipped.mean()
        if self.ppo2_vf_coef_halving:
            v_loss *= 0.5

        entropy_loss = -entropy.mean()

        loss = pi_loss + ent_coef * entropy_loss + self.vf_coef * v_loss

        self.optimizer.zero_grad()
        loss.backward()
        nn.utils.clip_grad_norm_(self.policy.parameters(), self.max_grad_norm)
        self.optimizer.step()

        with torch.no_grad():
            approx_kl = ((ratio - 1) - logratio).mean().cpu().numpy().item()
            clipped_frac = (
                ((ratio - 1).abs() > pi_clip).float().mean().cpu().numpy().item()
            )
            val_clipped_frac = (
                (((v - orig_v).abs() > v_clip).float().mean().cpu().numpy().item())
                if v_clip
                else 0
            )

        return TrainStepStats(
            loss.item(),
            pi_loss.item(),
            v_loss.item(),
            entropy_loss.item(),
            approx_kl,
            clipped_frac,
            val_clipped_frac,
        )