pushing model

.gitattributes

@@ -32,3 +32,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+videos/Atlantis-v5__sebulba_ppo_envpool_impala_atari_wrapper__2__fbe168b1-4c1d-4d09-be17-a0208c0f35e2-eval/0.mp4 filter=lfs diff=lfs merge=lfs -text
+replay.mp4 filter=lfs diff=lfs merge=lfs -text
+sebulba_ppo_envpool_impala_atari_wrapper.cleanrl_model filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,92 @@
+---
+tags:
+- Atlantis-v5
+- deep-reinforcement-learning
+- reinforcement-learning
+- custom-implementation
+library_name: cleanrl
+model-index:
+- name: PPO
+  results:
+  - task:
+      type: reinforcement-learning
+      name: reinforcement-learning
+    dataset:
+      name: Atlantis-v5
+      type: Atlantis-v5
+    metrics:
+    - type: mean_reward
+      value: 919490.00 +/- 10446.96
+      name: mean_reward
+      verified: false
+---
+
+# (CleanRL) **PPO** Agent Playing **Atlantis-v5**
+
+This is a trained model of a PPO agent playing Atlantis-v5.
+The model was trained by using [CleanRL](https://github.com/vwxyzjn/cleanrl) and the most up-to-date training code can be
+found [here](https://github.com/vwxyzjn/cleanrl/blob/master/cleanrl/sebulba_ppo_envpool_impala_atari_wrapper.py).
+
+## Get Started
+
+To use this model, please install the `cleanrl` package with the following command:
+
+```
+pip install "cleanrl[jax,envpool,atari]"
+python -m cleanrl_utils.enjoy --exp-name sebulba_ppo_envpool_impala_atari_wrapper --env-id Atlantis-v5
+```
+
+Please refer to the [documentation](https://docs.cleanrl.dev/get-started/zoo/) for more detail.
+
+
+## Command to reproduce the training
+
+```bash
+curl -OL https://huggingface.co/cleanrl/Atlantis-v5-sebulba_ppo_envpool_impala_atari_wrapper-seed2/raw/main/sebulba_ppo_envpool_impala_atari_wrapper.py
+curl -OL https://huggingface.co/cleanrl/Atlantis-v5-sebulba_ppo_envpool_impala_atari_wrapper-seed2/raw/main/pyproject.toml
+curl -OL https://huggingface.co/cleanrl/Atlantis-v5-sebulba_ppo_envpool_impala_atari_wrapper-seed2/raw/main/poetry.lock
+poetry install --all-extras
+python sebulba_ppo_envpool_impala_atari_wrapper.py --actor-device-ids 0 --learner-device-ids 1 2 3 4 5 6 --track --save-model --upload-model --hf-entity cleanrl --env-id Atlantis-v5 --seed 2
+```
+
+# Hyperparameters
+```python
+{'actor_device_ids': [0],
+ 'anneal_lr': True,
+ 'async_batch_size': 20,
+ 'async_update': 3,
+ 'batch_size': 7680,
+ 'capture_video': False,
+ 'clip_coef': 0.1,
+ 'cuda': True,
+ 'ent_coef': 0.01,
+ 'env_id': 'Atlantis-v5',
+ 'exp_name': 'sebulba_ppo_envpool_impala_atari_wrapper',
+ 'gae_lambda': 0.95,
+ 'gamma': 0.99,
+ 'hf_entity': 'cleanrl',
+ 'learner_device_ids': [1, 2, 3, 4, 5, 6],
+ 'learning_rate': 0.00025,
+ 'max_grad_norm': 0.5,
+ 'minibatch_size': 1920,
+ 'norm_adv': True,
+ 'num_actor_threads': 1,
+ 'num_envs': 60,
+ 'num_minibatches': 4,
+ 'num_steps': 128,
+ 'num_updates': 6510,
+ 'profile': False,
+ 'save_model': True,
+ 'seed': 2,
+ 'target_kl': None,
+ 'test_actor_learner_throughput': False,
+ 'torch_deterministic': True,
+ 'total_timesteps': 50000000,
+ 'track': True,
+ 'update_epochs': 4,
+ 'upload_model': True,
+ 'vf_coef': 0.5,
+ 'wandb_entity': None,
+ 'wandb_project_name': 'cleanRL'}
+```
+    

events.out.tfevents.1676001998.ip-26-0-128-237.393923.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:433f51c6e70263bdce49f7beec57b2b18a629c25d5dc7897408393cbe9c48aa5
+size 9506854

pyproject.toml

@@ -0,0 +1,178 @@
+[tool.poetry]
+name = "cleanrl"
+version = "1.1.0"
+description = "High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features"
+authors = ["Costa Huang <costa.huang@outlook.com>"]
+packages = [
+    { include = "cleanrl" },
+    { include = "cleanrl_utils" },
+]
+keywords = ["reinforcement", "machine", "learning", "research"]
+license="MIT"
+readme = "README.md"
+
+[tool.poetry.dependencies]
+python = ">=3.7.1,<3.10"
+tensorboard = "^2.10.0"
+wandb = "^0.13.6"
+gym = "0.23.1"
+torch = ">=1.12.1"
+stable-baselines3 = "1.2.0"
+gymnasium = "^0.26.3"
+moviepy = "^1.0.3"
+pygame = "2.1.0"
+huggingface-hub = "^0.11.1"
+
+ale-py = {version = "0.7.4", optional = true}
+AutoROM = {extras = ["accept-rom-license"], version = "^0.4.2"}
+opencv-python = {version = "^4.6.0.66", optional = true}
+pybullet = {version = "3.1.8", optional = true}
+procgen = {version = "^0.10.7", optional = true}
+pytest = {version = "^7.1.3", optional = true}
+mujoco = {version = "^2.2", optional = true}
+imageio = {version = "^2.14.1", optional = true}
+free-mujoco-py = {version = "^2.1.6", optional = true}
+mkdocs-material = {version = "^8.4.3", optional = true}
+markdown-include = {version = "^0.7.0", optional = true}
+jax = {version = "^0.3.17", optional = true}
+jaxlib = {version = "^0.3.15", optional = true}
+flax = {version = "^0.6.0", optional = true}
+optuna = {version = "^3.0.1", optional = true}
+optuna-dashboard = {version = "^0.7.2", optional = true}
+rich = {version = "<12.0", optional = true}
+envpool = {version = "^0.8.1", optional = true}
+PettingZoo = {version = "1.18.1", optional = true}
+SuperSuit = {version = "3.4.0", optional = true}
+multi-agent-ale-py = {version = "0.1.11", optional = true}
+boto3 = {version = "^1.24.70", optional = true}
+awscli = {version = "^1.25.71", optional = true}
+shimmy = {version = "^0.1.0", optional = true}
+dm-control = {version = "^1.0.8", optional = true}
+
+[tool.poetry.group.dev.dependencies]
+pre-commit = "^2.20.0"
+
+[tool.poetry.group.atari]
+optional = true
+[tool.poetry.group.atari.dependencies]
+ale-py = "0.7.4"
+AutoROM = {extras = ["accept-rom-license"], version = "^0.4.2"}
+opencv-python = "^4.6.0.66"
+
+[tool.poetry.group.pybullet]
+optional = true
+[tool.poetry.group.pybullet.dependencies]
+pybullet = "3.1.8"
+
+[tool.poetry.group.procgen]
+optional = true
+[tool.poetry.group.procgen.dependencies]
+procgen = "^0.10.7"
+
+[tool.poetry.group.pytest]
+optional = true
+[tool.poetry.group.pytest.dependencies]
+pytest = "^7.1.3"
+
+[tool.poetry.group.mujoco]
+optional = true
+[tool.poetry.group.mujoco.dependencies]
+mujoco = "^2.2"
+imageio = "^2.14.1"
+
+[tool.poetry.group.mujoco_py]
+optional = true
+[tool.poetry.group.mujoco_py.dependencies]
+free-mujoco-py = "^2.1.6"
+
+[tool.poetry.group.docs]
+optional = true
+[tool.poetry.group.docs.dependencies]
+mkdocs-material = "^8.4.3"
+markdown-include = "^0.7.0"
+
+[tool.poetry.group.jax]
+optional = true
+[tool.poetry.group.jax.dependencies]
+jax = "^0.3.17"
+jaxlib = "^0.3.15"
+flax = "^0.6.0"
+
+[tool.poetry.group.optuna]
+optional = true
+[tool.poetry.group.optuna.dependencies]
+optuna = "^3.0.1"
+optuna-dashboard = "^0.7.2"
+rich = "<12.0"
+
+[tool.poetry.group.envpool]
+optional = true
+[tool.poetry.group.envpool.dependencies]
+envpool = "^0.8.1"
+
+[tool.poetry.group.pettingzoo]
+optional = true
+[tool.poetry.group.pettingzoo.dependencies]
+PettingZoo = "1.18.1"
+SuperSuit = "3.4.0"
+multi-agent-ale-py = "0.1.11"
+
+[tool.poetry.group.cloud]
+optional = true
+[tool.poetry.group.cloud.dependencies]
+boto3 = "^1.24.70"
+awscli = "^1.25.71"
+
+[tool.poetry.group.isaacgym]
+optional = true
+[tool.poetry.group.isaacgym.dependencies]
+isaacgymenvs = {git = "https://github.com/vwxyzjn/IsaacGymEnvs.git", rev = "poetry"}
+isaacgym = {path = "cleanrl/ppo_continuous_action_isaacgym/isaacgym", develop = true}
+
+[tool.poetry.group.dm_control]
+optional = true
+[tool.poetry.group.dm_control.dependencies]
+shimmy = "^0.1.0"
+dm-control = "^1.0.8"
+mujoco = "^2.2"
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
+
+[tool.poetry.extras]
+atari = ["ale-py", "AutoROM", "opencv-python"]
+pybullet = ["pybullet"]
+procgen = ["procgen"]
+plot = ["pandas", "seaborn"]
+pytest = ["pytest"]
+mujoco = ["mujoco", "imageio"]
+mujoco_py = ["free-mujoco-py"]
+jax = ["jax", "jaxlib", "flax"]
+docs = ["mkdocs-material", "markdown-include"]
+envpool = ["envpool"]
+optuna = ["optuna", "optuna-dashboard", "rich"]
+pettingzoo = ["PettingZoo", "SuperSuit", "multi-agent-ale-py"]
+cloud = ["boto3", "awscli"]
+dm_control = ["shimmy", "dm-control", "mujoco"]
+
+# dependencies for algorithm variant (useful when you want to run a specific algorithm)
+dqn = []
+dqn_atari = ["ale-py", "AutoROM", "opencv-python"]
+dqn_jax = ["jax", "jaxlib", "flax"]
+dqn_atari_jax = [
+    "ale-py", "AutoROM", "opencv-python", # atari
+    "jax", "jaxlib", "flax" # jax
+]
+c51 = []
+c51_atari = ["ale-py", "AutoROM", "opencv-python"]
+c51_jax = ["jax", "jaxlib", "flax"]
+c51_atari_jax = [
+    "ale-py", "AutoROM", "opencv-python", # atari
+    "jax", "jaxlib", "flax" # jax
+]
+ppo_atari_envpool_xla_jax_scan = [
+    "ale-py", "AutoROM", "opencv-python", # atari
+    "jax", "jaxlib", "flax", # jax
+    "envpool", # envpool
+]

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b844277df3361670765580dc1b1d3380bd55cd52f62c2dadc2264a43cdad900
+size 1771674

sebulba_ppo_envpool_impala_atari_wrapper.cleanrl_model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8635f84cdd58dfe098fc05712d591b52c417e5c7e890f87f9de1f44186987691
+size 4363948

sebulba_ppo_envpool_impala_atari_wrapper.py

@@ -0,0 +1,835 @@
+"""
+* 🥼 Test throughput (see docs):
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 1 --profile --test-actor-learner-throughput --total-timesteps 500000 --track
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 1 2 --profile --test-actor-learner-throughput --total-timesteps 500000 --track
+        * this will help us diagnose the throughput issue
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --num-actor-threads 2 --learner-device-ids 1 --profile --test-actor-learner-throughput --total-timesteps 500000 --track
+* 🔥 Best performance so far (more GPUs -> faster)
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 0 --track
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 0 1 --track
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 1 2 3 --track
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 1 2 3 4 --track
+    * python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 1 2 3 4 5 6 --track
+    * (this actually doesn't work that well) python sebulba_ppo_envpool.py --actor-device-ids 0 --learner-device-ids 1 2 3 4 5 6 7 --num-envs 70 --async-batch-size 35 --track
+"""
+# docs and experiment results can be found at https://docs.cleanrl.dev/rl-algorithms/ppo/#ppo_atari_envpool_async_jax_scan_impalanet_machadopy
+import argparse
+import os
+import random
+import time
+import uuid
+import warnings
+from collections import deque
+from distutils.util import strtobool
+from functools import partial
+from typing import Sequence
+
+os.environ[
+    "XLA_PYTHON_CLIENT_MEM_FRACTION"
+] = "0.6"  # see https://github.com/google/jax/discussions/6332#discussioncomment-1279991
+os.environ["XLA_FLAGS"] = "--xla_cpu_multi_thread_eigen=false " "intra_op_parallelism_threads=1"
+import multiprocessing as mp
+import queue
+import threading
+
+import envpool
+import flax
+import flax.linen as nn
+import gym
+import jax
+import jax.numpy as jnp
+import numpy as np
+import optax
+from flax.linen.initializers import constant, orthogonal
+from flax.training.train_state import TrainState
+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="weather 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="Breakout-v5",
+        help="the id of the environment")
+    parser.add_argument("--total-timesteps", type=int, default=50000000,
+        help="total timesteps of the experiments")
+    parser.add_argument("--learning-rate", type=float, default=2.5e-4,
+        help="the learning rate of the optimizer")
+    parser.add_argument("--num-envs", type=int, default=60,
+        help="the number of parallel game environments")
+    parser.add_argument("--async-batch-size", type=int, default=20,
+        help="the envpool's batch size in the async mode")
+    parser.add_argument("--num-steps", type=int, default=128,
+        help="the number of steps to run in each environment per policy rollout")
+    parser.add_argument("--anneal-lr", type=lambda x: bool(strtobool(x)), default=True, nargs="?", const=True,
+        help="Toggle learning rate annealing for policy and value networks")
+    parser.add_argument("--gamma", type=float, default=0.99,
+        help="the discount factor gamma")
+    parser.add_argument("--gae-lambda", type=float, default=0.95,
+        help="the lambda for the general advantage estimation")
+    parser.add_argument("--num-minibatches", type=int, default=4,
+        help="the number of mini-batches")
+    parser.add_argument("--update-epochs", type=int, default=4,
+        help="the K epochs to update the policy")
+    parser.add_argument("--norm-adv", type=lambda x: bool(strtobool(x)), default=True, nargs="?", const=True,
+        help="Toggles advantages normalization")
+    parser.add_argument("--clip-coef", type=float, default=0.1,
+        help="the surrogate clipping coefficient")
+    parser.add_argument("--ent-coef", type=float, default=0.01,
+        help="coefficient of the entropy")
+    parser.add_argument("--vf-coef", type=float, default=0.5,
+        help="coefficient of the value function")
+    parser.add_argument("--max-grad-norm", type=float, default=0.5,
+        help="the maximum norm for the gradient clipping")
+    parser.add_argument("--target-kl", type=float, default=None,
+        help="the target KL divergence threshold")
+
+    parser.add_argument("--actor-device-ids", type=int, nargs="+", default=[0], # type is actually List[int]
+        help="the device ids that actor workers will use")
+    parser.add_argument("--learner-device-ids", type=int, nargs="+", default=[0], # type is actually List[int]
+        help="the device ids that actor workers will use")
+    parser.add_argument("--num-actor-threads", type=int, default=1,
+        help="the number of actor threads")
+    parser.add_argument("--profile", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
+        help="whether to call block_until_ready() for profiling")
+    parser.add_argument("--test-actor-learner-throughput", type=lambda x: bool(strtobool(x)), default=False, nargs="?", const=True,
+        help="whether to test actor-learner throughput by removing the actor-learner communication")
+    args = parser.parse_args()
+    args.batch_size = int(args.num_envs * args.num_steps)
+    args.minibatch_size = int(args.batch_size // args.num_minibatches)
+    args.num_updates = args.total_timesteps // args.batch_size
+    args.async_update = int(args.num_envs / args.async_batch_size)
+    assert len(args.actor_device_ids) == 1, "only 1 actor_device_ids is supported now"
+    if args.num_actor_threads > 1:
+        warnings.warn("⚠️ !!!! `num_actor_threads` > 1 is not tested with learning; see docs for detail")
+    # fmt: on
+    return args
+
+
+ATARI_MAX_FRAMES = int(
+    108000 / 4
+)  # 108000 is the max number of frames in an Atari game, divided by 4 to account for frame skipping
+
+
+def make_env(env_id, seed, num_envs, async_batch_size=1, num_threads=None, thread_affinity_offset=-1):
+    def thunk():
+        envs = envpool.make(
+            env_id,
+            env_type="gym",
+            num_envs=num_envs,
+            num_threads=num_threads if num_threads is not None else async_batch_size,
+            thread_affinity_offset=thread_affinity_offset,
+            batch_size=async_batch_size,
+            episodic_life=True,  # Espeholt et al., 2018, Tab. G.1
+            repeat_action_probability=0,  # Hessel et al., 2022 (Muesli) Tab. 10
+            noop_max=30,  # Espeholt et al., 2018, Tab. C.1 "Up to 30 no-ops at the beginning of each episode."
+            full_action_space=False,  # Espeholt et al., 2018, Appendix G., "Following related work, experts use game-specific action sets."
+            max_episode_steps=ATARI_MAX_FRAMES,  # Hessel et al. 2018 (Rainbow DQN), Table 3, Max frames per episode
+            reward_clip=True,
+            seed=seed,
+        )
+        envs.num_envs = num_envs
+        envs.single_action_space = envs.action_space
+        envs.single_observation_space = envs.observation_space
+        envs.is_vector_env = True
+        return envs
+
+    return thunk
+
+
+class ResidualBlock(nn.Module):
+    channels: int
+
+    @nn.compact
+    def __call__(self, x):
+        inputs = x
+        x = nn.relu(x)
+        x = nn.Conv(
+            self.channels,
+            kernel_size=(3, 3),
+        )(x)
+        x = nn.relu(x)
+        x = nn.Conv(
+            self.channels,
+            kernel_size=(3, 3),
+        )(x)
+        return x + inputs
+
+
+class ConvSequence(nn.Module):
+    channels: int
+
+    @nn.compact
+    def __call__(self, x):
+        x = nn.Conv(
+            self.channels,
+            kernel_size=(3, 3),
+        )(x)
+        x = nn.max_pool(x, window_shape=(3, 3), strides=(2, 2), padding="SAME")
+        x = ResidualBlock(self.channels)(x)
+        x = ResidualBlock(self.channels)(x)
+        return x
+
+
+class Network(nn.Module):
+    channelss: Sequence[int] = (16, 32, 32)
+
+    @nn.compact
+    def __call__(self, x):
+        x = jnp.transpose(x, (0, 2, 3, 1))
+        x = x / (255.0)
+        for channels in self.channelss:
+            x = ConvSequence(channels)(x)
+        x = nn.relu(x)
+        x = x.reshape((x.shape[0], -1))
+        x = nn.Dense(256, kernel_init=orthogonal(np.sqrt(2)), bias_init=constant(0.0))(x)
+        x = nn.relu(x)
+        return x
+
+
+class Critic(nn.Module):
+    @nn.compact
+    def __call__(self, x):
+        return nn.Dense(1, kernel_init=orthogonal(1), bias_init=constant(0.0))(x)
+
+
+class Actor(nn.Module):
+    action_dim: int
+
+    @nn.compact
+    def __call__(self, x):
+        return nn.Dense(self.action_dim, kernel_init=orthogonal(0.01), bias_init=constant(0.0))(x)
+
+
+@flax.struct.dataclass
+class AgentParams:
+    network_params: flax.core.FrozenDict
+    actor_params: flax.core.FrozenDict
+    critic_params: flax.core.FrozenDict
+
+
+@partial(jax.jit, static_argnums=(3))
+def get_action_and_value(
+    params: TrainState,
+    next_obs: np.ndarray,
+    key: jax.random.PRNGKey,
+    action_dim: int,
+):
+    hidden = Network().apply(params.network_params, next_obs)
+    logits = Actor(action_dim).apply(params.actor_params, hidden)
+    # sample action: Gumbel-softmax trick
+    # see https://stats.stackexchange.com/questions/359442/sampling-from-a-categorical-distribution
+    key, subkey = jax.random.split(key)
+    u = jax.random.uniform(subkey, shape=logits.shape)
+    action = jnp.argmax(logits - jnp.log(-jnp.log(u)), axis=1)
+    logprob = jax.nn.log_softmax(logits)[jnp.arange(action.shape[0]), action]
+    value = Critic().apply(params.critic_params, hidden)
+    return action, logprob, value.squeeze(), key
+
+
+@jax.jit
+def prepare_data(
+    obs: list,
+    dones: list,
+    values: list,
+    actions: list,
+    logprobs: list,
+    env_ids: list,
+    rewards: list,
+):
+    obs = jnp.asarray(obs)
+    dones = jnp.asarray(dones)
+    values = jnp.asarray(values)
+    actions = jnp.asarray(actions)
+    logprobs = jnp.asarray(logprobs)
+    env_ids = jnp.asarray(env_ids)
+    rewards = jnp.asarray(rewards)
+
+    # TODO: in an unlikely event, one of the envs might have not stepped at all, which may results in unexpected behavior
+    T, B = env_ids.shape
+    index_ranges = jnp.arange(T * B, dtype=jnp.int32)
+    next_index_ranges = jnp.zeros_like(index_ranges, dtype=jnp.int32)
+    last_env_ids = jnp.zeros(args.num_envs, dtype=jnp.int32) - 1
+
+    def f(carry, x):
+        last_env_ids, next_index_ranges = carry
+        env_id, index_range = x
+        next_index_ranges = next_index_ranges.at[last_env_ids[env_id]].set(
+            jnp.where(last_env_ids[env_id] != -1, index_range, next_index_ranges[last_env_ids[env_id]])
+        )
+        last_env_ids = last_env_ids.at[env_id].set(index_range)
+        return (last_env_ids, next_index_ranges), None
+
+    (last_env_ids, next_index_ranges), _ = jax.lax.scan(
+        f,
+        (last_env_ids, next_index_ranges),
+        (env_ids.reshape(-1), index_ranges),
+    )
+
+    # rewards is off by one time step
+    rewards = rewards.reshape(-1)[next_index_ranges].reshape((args.num_steps) * args.async_update, args.async_batch_size)
+    advantages, returns, _, final_env_ids = compute_gae(env_ids, rewards, values, dones)
+    # b_inds = jnp.nonzero(final_env_ids.reshape(-1), size=(args.num_steps) * args.async_update * args.async_batch_size)[0] # useful for debugging
+    b_obs = obs.reshape((-1,) + obs.shape[2:])
+    b_actions = actions.reshape(-1)
+    b_logprobs = logprobs.reshape(-1)
+    b_advantages = advantages.reshape(-1)
+    b_returns = returns.reshape(-1)
+    return b_obs, b_actions, b_logprobs, b_advantages, b_returns
+
+
+def rollout(
+    i,
+    num_threads,  # =None,
+    thread_affinity_offset,  # =-1,
+    key: jax.random.PRNGKey,
+    args,
+    rollout_queue,
+    params_queue: queue.Queue,
+    writer,
+    learner_devices,
+):
+    envs = make_env(args.env_id, args.seed, args.num_envs, args.async_batch_size, num_threads, thread_affinity_offset)()
+    len_actor_device_ids = len(args.actor_device_ids)
+    global_step = 0
+    # TRY NOT TO MODIFY: start the game
+    start_time = time.time()
+
+    # put data in the last index
+    episode_returns = np.zeros((args.num_envs,), dtype=np.float32)
+    returned_episode_returns = np.zeros((args.num_envs,), dtype=np.float32)
+    episode_lengths = np.zeros((args.num_envs,), dtype=np.float32)
+    returned_episode_lengths = np.zeros((args.num_envs,), dtype=np.float32)
+    envs.async_reset()
+
+    params_queue_get_time = deque(maxlen=10)
+    rollout_time = deque(maxlen=10)
+    data_transfer_time = deque(maxlen=10)
+    rollout_queue_put_time = deque(maxlen=10)
+    actor_policy_version = 0
+    for update in range(1, args.num_updates + 2):
+        # NOTE: This is a major difference from the sync version:
+        # at the end of the rollout phase, the sync version will have the next observation
+        # ready for the value bootstrap, but the async version will not have it.
+        # for this reason we do `num_steps + 1`` to get the extra states for value bootstrapping.
+        # but note that the extra states are not used for the loss computation in the next iteration,
+        # while the sync version will use the extra state for the loss computation.
+        update_time_start = time.time()
+        obs = []
+        dones = []
+        actions = []
+        logprobs = []
+        values = []
+        env_ids = []
+        rewards = []
+        truncations = []
+        terminations = []
+        env_recv_time = 0
+        inference_time = 0
+        storage_time = 0
+        env_send_time = 0
+
+        # NOTE: `update != 2` is actually IMPORTANT — it allows us to start running policy collection
+        # concurrently with the learning process. It also ensures the actor's policy version is only 1 step
+        # behind the learner's policy version
+        params_queue_get_time_start = time.time()
+        if update != 2:
+            params = params_queue.get()
+            actor_policy_version += 1
+        params_queue_get_time.append(time.time() - params_queue_get_time_start)
+        writer.add_scalar("stats/params_queue_get_time", np.mean(params_queue_get_time), global_step)
+        rollout_time_start = time.time()
+        for _ in range(
+            args.async_update, (args.num_steps + 1) * args.async_update
+        ):  # num_steps + 1 to get the states for value bootstrapping.
+            env_recv_time_start = time.time()
+            next_obs, next_reward, next_done, info = envs.recv()
+            env_recv_time += time.time() - env_recv_time_start
+            global_step += len(next_done) * args.num_actor_threads * len_actor_device_ids
+            env_id = info["env_id"]
+
+            inference_time_start = time.time()
+            action, logprob, value, key = get_action_and_value(params, next_obs, key, envs.single_action_space.n)
+            inference_time += time.time() - inference_time_start
+
+            env_send_time_start = time.time()
+            envs.send(np.array(action), env_id)
+            env_send_time += time.time() - env_send_time_start
+            storage_time_start = time.time()
+            obs.append(next_obs)
+            dones.append(next_done)
+            values.append(value)
+            actions.append(action)
+            logprobs.append(logprob)
+            env_ids.append(env_id)
+            rewards.append(next_reward)
+
+            # info["TimeLimit.truncated"] has a bug https://github.com/sail-sg/envpool/issues/239
+            # so we use our own truncated flag
+            truncated = info["elapsed_step"] >= ATARI_MAX_FRAMES
+            truncations.append(truncated)
+            terminations.append(info["terminated"])
+            episode_returns[env_id] += info["reward"]
+            returned_episode_returns[env_id] = np.where(
+                info["terminated"] + truncated, episode_returns[env_id], returned_episode_returns[env_id]
+            )
+            episode_returns[env_id] *= (1 - info["terminated"]) * (1 - truncated)
+            episode_lengths[env_id] += 1
+            returned_episode_lengths[env_id] = np.where(
+                info["terminated"] + truncated, episode_lengths[env_id], returned_episode_lengths[env_id]
+            )
+            episode_lengths[env_id] *= (1 - info["terminated"]) * (1 - truncated)
+            storage_time += time.time() - storage_time_start
+        if args.profile:
+            action.block_until_ready()
+        rollout_time.append(time.time() - rollout_time_start)
+        writer.add_scalar("stats/rollout_time", np.mean(rollout_time), global_step)
+
+        avg_episodic_return = np.mean(returned_episode_returns)
+        writer.add_scalar("charts/avg_episodic_return", avg_episodic_return, global_step)
+        writer.add_scalar("charts/avg_episodic_length", np.mean(returned_episode_lengths), global_step)
+        if i == 0:
+            print(f"global_step={global_step}, avg_episodic_return={avg_episodic_return}")
+            print("SPS:", int(global_step / (time.time() - start_time)))
+        writer.add_scalar("charts/SPS", int(global_step / (time.time() - start_time)), global_step)
+
+        writer.add_scalar("stats/truncations", np.sum(truncations), global_step)
+        writer.add_scalar("stats/terminations", np.sum(terminations), global_step)
+        writer.add_scalar("stats/env_recv_time", env_recv_time, global_step)
+        writer.add_scalar("stats/inference_time", inference_time, global_step)
+        writer.add_scalar("stats/storage_time", storage_time, global_step)
+        writer.add_scalar("stats/env_send_time", env_send_time, global_step)
+
+        data_transfer_time_start = time.time()
+        b_obs, b_actions, b_logprobs, b_advantages, b_returns = prepare_data(
+            obs,
+            dones,
+            values,
+            actions,
+            logprobs,
+            env_ids,
+            rewards,
+        )
+        payload = (
+            global_step,
+            actor_policy_version,
+            update,
+            jnp.array_split(b_obs, len(learner_devices)),
+            jnp.array_split(b_actions, len(learner_devices)),
+            jnp.array_split(b_logprobs, len(learner_devices)),
+            jnp.array_split(b_advantages, len(learner_devices)),
+            jnp.array_split(b_returns, len(learner_devices)),
+        )
+        if args.profile:
+            payload[2][0].block_until_ready()
+        data_transfer_time.append(time.time() - data_transfer_time_start)
+        writer.add_scalar("stats/data_transfer_time", np.mean(data_transfer_time), global_step)
+        if update == 1 or not args.test_actor_learner_throughput:
+            rollout_queue_put_time_start = time.time()
+            rollout_queue.put(payload)
+            rollout_queue_put_time.append(time.time() - rollout_queue_put_time_start)
+            writer.add_scalar("stats/rollout_queue_put_time", np.mean(rollout_queue_put_time), global_step)
+
+        writer.add_scalar(
+            "charts/SPS_update",
+            int(
+                args.num_envs
+                * args.num_steps
+                * args.num_actor_threads
+                * len_actor_device_ids
+                / (time.time() - update_time_start)
+            ),
+            global_step,
+        )
+
+
+@partial(jax.jit, static_argnums=(3))
+def get_action_and_value2(
+    params: flax.core.FrozenDict,
+    x: np.ndarray,
+    action: np.ndarray,
+    action_dim: int,
+):
+    hidden = Network().apply(params.network_params, x)
+    logits = Actor(action_dim).apply(params.actor_params, hidden)
+    logprob = jax.nn.log_softmax(logits)[jnp.arange(action.shape[0]), action]
+    logits = logits - jax.scipy.special.logsumexp(logits, axis=-1, keepdims=True)
+    logits = logits.clip(min=jnp.finfo(logits.dtype).min)
+    p_log_p = logits * jax.nn.softmax(logits)
+    entropy = -p_log_p.sum(-1)
+    value = Critic().apply(params.critic_params, hidden).squeeze()
+    return logprob, entropy, value
+
+
+@jax.jit
+def compute_gae(
+    env_ids: np.ndarray,
+    rewards: np.ndarray,
+    values: np.ndarray,
+    dones: np.ndarray,
+):
+    dones = jnp.asarray(dones)
+    values = jnp.asarray(values)
+    env_ids = jnp.asarray(env_ids)
+    rewards = jnp.asarray(rewards)
+
+    _, B = env_ids.shape
+    final_env_id_checked = jnp.zeros(args.num_envs, jnp.int32) - 1
+    final_env_ids = jnp.zeros(B, jnp.int32)
+    advantages = jnp.zeros(B)
+    lastgaelam = jnp.zeros(args.num_envs)
+    lastdones = jnp.zeros(args.num_envs) + 1
+    lastvalues = jnp.zeros(args.num_envs)
+
+    def compute_gae_once(carry, x):
+        lastvalues, lastdones, advantages, lastgaelam, final_env_ids, final_env_id_checked = carry
+        (
+            done,
+            value,
+            eid,
+            reward,
+        ) = x
+        nextnonterminal = 1.0 - lastdones[eid]
+        nextvalues = lastvalues[eid]
+        delta = jnp.where(final_env_id_checked[eid] == -1, 0, reward + args.gamma * nextvalues * nextnonterminal - value)
+        advantages = delta + args.gamma * args.gae_lambda * nextnonterminal * lastgaelam[eid]
+        final_env_ids = jnp.where(final_env_id_checked[eid] == 1, 1, 0)
+        final_env_id_checked = final_env_id_checked.at[eid].set(
+            jnp.where(final_env_id_checked[eid] == -1, 1, final_env_id_checked[eid])
+        )
+
+        # the last_ variables keeps track of the actual `num_steps`
+        lastgaelam = lastgaelam.at[eid].set(advantages)
+        lastdones = lastdones.at[eid].set(done)
+        lastvalues = lastvalues.at[eid].set(value)
+        return (lastvalues, lastdones, advantages, lastgaelam, final_env_ids, final_env_id_checked), (
+            advantages,
+            final_env_ids,
+        )
+
+    (_, _, _, _, final_env_ids, final_env_id_checked), (advantages, final_env_ids) = jax.lax.scan(
+        compute_gae_once,
+        (
+            lastvalues,
+            lastdones,
+            advantages,
+            lastgaelam,
+            final_env_ids,
+            final_env_id_checked,
+        ),
+        (
+            dones,
+            values,
+            env_ids,
+            rewards,
+        ),
+        reverse=True,
+    )
+    return advantages, advantages + values, final_env_id_checked, final_env_ids
+
+
+def ppo_loss(params, x, a, logp, mb_advantages, mb_returns, action_dim):
+    newlogprob, entropy, newvalue = get_action_and_value2(params, x, a, action_dim)
+    logratio = newlogprob - logp
+    ratio = jnp.exp(logratio)
+    approx_kl = ((ratio - 1) - logratio).mean()
+
+    if args.norm_adv:
+        mb_advantages = (mb_advantages - mb_advantages.mean()) / (mb_advantages.std() + 1e-8)
+
+    # Policy loss
+    pg_loss1 = -mb_advantages * ratio
+    pg_loss2 = -mb_advantages * jnp.clip(ratio, 1 - args.clip_coef, 1 + args.clip_coef)
+    pg_loss = jnp.maximum(pg_loss1, pg_loss2).mean()
+
+    # Value loss
+    v_loss = 0.5 * ((newvalue - mb_returns) ** 2).mean()
+
+    entropy_loss = entropy.mean()
+    loss = pg_loss - args.ent_coef * entropy_loss + v_loss * args.vf_coef
+    return loss, (pg_loss, v_loss, entropy_loss, jax.lax.stop_gradient(approx_kl))
+
+
+@partial(jax.jit, static_argnums=(6))
+def single_device_update(
+    agent_state: TrainState,
+    b_obs,
+    b_actions,
+    b_logprobs,
+    b_advantages,
+    b_returns,
+    action_dim,
+    key: jax.random.PRNGKey,
+):
+    ppo_loss_grad_fn = jax.value_and_grad(ppo_loss, has_aux=True)
+
+    def update_epoch(carry, _):
+        agent_state, key = carry
+        key, subkey = jax.random.split(key)
+
+        # taken from: https://github.com/google/brax/blob/main/brax/training/agents/ppo/train.py
+        def convert_data(x: jnp.ndarray):
+            x = jax.random.permutation(subkey, x)
+            x = jnp.reshape(x, (args.num_minibatches, -1) + x.shape[1:])
+            return x
+
+        def update_minibatch(agent_state, minibatch):
+            mb_obs, mb_actions, mb_logprobs, mb_advantages, mb_returns = minibatch
+            (loss, (pg_loss, v_loss, entropy_loss, approx_kl)), grads = ppo_loss_grad_fn(
+                agent_state.params,
+                mb_obs,
+                mb_actions,
+                mb_logprobs,
+                mb_advantages,
+                mb_returns,
+                action_dim,
+            )
+            grads = jax.lax.pmean(grads, axis_name="devices")
+            agent_state = agent_state.apply_gradients(grads=grads)
+            return agent_state, (loss, pg_loss, v_loss, entropy_loss, approx_kl, grads)
+
+        agent_state, (loss, pg_loss, v_loss, entropy_loss, approx_kl, grads) = jax.lax.scan(
+            update_minibatch,
+            agent_state,
+            (
+                convert_data(b_obs),
+                convert_data(b_actions),
+                convert_data(b_logprobs),
+                convert_data(b_advantages),
+                convert_data(b_returns),
+            ),
+        )
+        return (agent_state, key), (loss, pg_loss, v_loss, entropy_loss, approx_kl, grads)
+
+    (agent_state, key), (loss, pg_loss, v_loss, entropy_loss, approx_kl, _) = jax.lax.scan(
+        update_epoch, (agent_state, key), (), length=args.update_epochs
+    )
+    return agent_state, loss, pg_loss, v_loss, entropy_loss, approx_kl, key
+
+
+if __name__ == "__main__":
+    devices = jax.devices("gpu")
+    args = parse_args()
+    run_name = f"{args.env_id}__{args.exp_name}__{args.seed}__{uuid.uuid4()}"
+    if args.track:
+        import wandb
+
+        wandb.init(
+            project=args.wandb_project_name,
+            entity=args.wandb_entity,
+            sync_tensorboard=True,
+            config=vars(args),
+            name=run_name,
+            monitor_gym=True,
+            save_code=True,
+        )
+    print(devices)
+    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()])),
+    )
+
+    # TRY NOT TO MODIFY: seeding
+    random.seed(args.seed)
+    np.random.seed(args.seed)
+    key = jax.random.PRNGKey(args.seed)
+    key, network_key, actor_key, critic_key = jax.random.split(key, 4)
+
+    # env setup
+    envs = make_env(args.env_id, args.seed, args.num_envs, args.async_batch_size)()
+    assert isinstance(envs.single_action_space, gym.spaces.Discrete), "only discrete action space is supported"
+
+    def linear_schedule(count):
+        # anneal learning rate linearly after one training iteration which contains
+        # (args.num_minibatches * args.update_epochs) gradient updates
+        frac = 1.0 - (count // (args.num_minibatches * args.update_epochs)) / args.num_updates
+        return args.learning_rate * frac
+
+    network = Network()
+    actor = Actor(action_dim=envs.single_action_space.n)
+    critic = Critic()
+    network_params = network.init(network_key, np.array([envs.single_observation_space.sample()]))
+    agent_state = TrainState.create(
+        apply_fn=None,
+        params=AgentParams(
+            network_params,
+            actor.init(actor_key, network.apply(network_params, np.array([envs.single_observation_space.sample()]))),
+            critic.init(critic_key, network.apply(network_params, np.array([envs.single_observation_space.sample()]))),
+        ),
+        tx=optax.chain(
+            optax.clip_by_global_norm(args.max_grad_norm),
+            optax.inject_hyperparams(optax.adam)(
+                learning_rate=linear_schedule if args.anneal_lr else args.learning_rate, eps=1e-5
+            ),
+        ),
+    )
+    learner_devices = [devices[d_id] for d_id in args.learner_device_ids]
+    actor_devices = [devices[d_id] for d_id in args.actor_device_ids]
+    agent_state = flax.jax_utils.replicate(agent_state, devices=learner_devices)
+
+    multi_device_update = jax.pmap(
+        single_device_update,
+        axis_name="devices",
+        devices=learner_devices,
+        in_axes=(0, 0, 0, 0, 0, 0, None, None),
+        out_axes=(0, 0, 0, 0, 0, 0, None),
+        static_broadcasted_argnums=(6),
+    )
+
+    rollout_queue = queue.Queue(maxsize=1)
+    params_queues = []
+    num_cpus = mp.cpu_count()
+    fair_num_cpus = num_cpus // len(args.actor_device_ids)
+
+    class DummyWriter:
+        def add_scalar(self, arg0, arg1, arg3):
+            pass
+
+    dummy_writer = DummyWriter()
+    for d_idx, d_id in enumerate(args.actor_device_ids):
+        for j in range(args.num_actor_threads):
+            params_queue = queue.Queue(maxsize=1)
+            params_queue.put(jax.device_put(flax.jax_utils.unreplicate(agent_state.params), devices[d_id]))
+            threading.Thread(
+                target=rollout,
+                args=(
+                    j,
+                    fair_num_cpus if args.num_actor_threads > 1 else None,
+                    j * args.num_actor_threads if args.num_actor_threads > 1 else -1,
+                    jax.device_put(key, devices[d_id]),
+                    args,
+                    rollout_queue,
+                    params_queue,
+                    writer if d_idx == 0 and j == 0 else dummy_writer,
+                    learner_devices,
+                ),
+            ).start()
+            params_queues.append(params_queue)
+
+    rollout_queue_get_time = deque(maxlen=10)
+    learner_policy_version = 0
+    while True:
+        learner_policy_version += 1
+        if learner_policy_version == 1 or not args.test_actor_learner_throughput:
+            rollout_queue_get_time_start = time.time()
+            (
+                global_step,
+                actor_policy_version,
+                update,
+                b_obs,
+                b_actions,
+                b_logprobs,
+                b_advantages,
+                b_returns,
+            ) = rollout_queue.get()
+            rollout_queue_get_time.append(time.time() - rollout_queue_get_time_start)
+            writer.add_scalar("stats/rollout_queue_get_time", np.mean(rollout_queue_get_time), global_step)
+
+        training_time_start = time.time()
+        (agent_state, loss, pg_loss, v_loss, entropy_loss, approx_kl, key) = multi_device_update(
+            agent_state,
+            jax.device_put_sharded(b_obs, learner_devices),
+            jax.device_put_sharded(b_actions, learner_devices),
+            jax.device_put_sharded(b_logprobs, learner_devices),
+            jax.device_put_sharded(b_advantages, learner_devices),
+            jax.device_put_sharded(b_returns, learner_devices),
+            envs.single_action_space.n,
+            key,
+        )
+        if learner_policy_version == 1 or not args.test_actor_learner_throughput:
+            for d_idx, d_id in enumerate(args.actor_device_ids):
+                for j in range(args.num_actor_threads):
+                    params_queues[d_idx * args.num_actor_threads + j].put(
+                        jax.device_put(flax.jax_utils.unreplicate(agent_state.params), devices[d_id])
+                    )
+        if args.profile:
+            v_loss[-1, -1, -1].block_until_ready()
+        writer.add_scalar("stats/training_time", time.time() - training_time_start, global_step)
+        writer.add_scalar("stats/rollout_queue_size", rollout_queue.qsize(), global_step)
+        writer.add_scalar("stats/params_queue_size", params_queue.qsize(), global_step)
+        print(
+            global_step,
+            f"actor_policy_version={actor_policy_version}, actor_update={update}, learner_policy_version={learner_policy_version}, training time: {time.time() - training_time_start}s",
+        )
+
+        # TRY NOT TO MODIFY: record rewards for plotting purposes
+        writer.add_scalar("charts/learning_rate", agent_state.opt_state[1].hyperparams["learning_rate"][0].item(), global_step)
+        writer.add_scalar("losses/value_loss", v_loss[-1, -1, -1].item(), global_step)
+        writer.add_scalar("losses/policy_loss", pg_loss[-1, -1, -1].item(), global_step)
+        writer.add_scalar("losses/entropy", entropy_loss[-1, -1, -1].item(), global_step)
+        writer.add_scalar("losses/approx_kl", approx_kl[-1, -1, -1].item(), global_step)
+        writer.add_scalar("losses/loss", loss[-1, -1, -1].item(), global_step)
+        if update >= args.num_updates:
+            break
+
+    if args.save_model:
+        agent_state = flax.jax_utils.unreplicate(agent_state)
+        model_path = f"runs/{run_name}/{args.exp_name}.cleanrl_model"
+        with open(model_path, "wb") as f:
+            f.write(
+                flax.serialization.to_bytes(
+                    [
+                        vars(args),
+                        [
+                            agent_state.params.network_params,
+                            agent_state.params.actor_params,
+                            agent_state.params.critic_params,
+                        ],
+                    ]
+                )
+            )
+        print(f"model saved to {model_path}")
+        from cleanrl_utils.evals.ppo_envpool_jax_eval import evaluate
+
+        episodic_returns = evaluate(
+            model_path,
+            make_env,
+            args.env_id,
+            eval_episodes=10,
+            run_name=f"{run_name}-eval",
+            Model=(Network, Actor, Critic),
+        )
+        for idx, episodic_return in enumerate(episodic_returns):
+            writer.add_scalar("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,
+                episodic_returns,
+                repo_id,
+                "PPO",
+                f"runs/{run_name}",
+                f"videos/{run_name}-eval",
+                extra_dependencies=["jax", "envpool", "atari"],
+            )
+
+    envs.close()
+    writer.close()

videos/Atlantis-v5__sebulba_ppo_envpool_impala_atari_wrapper__2__fbe168b1-4c1d-4d09-be17-a0208c0f35e2-eval/0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b844277df3361670765580dc1b1d3380bd55cd52f62c2dadc2264a43cdad900
+size 1771674