Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Link CleanRL's PPO Implementation #2780

Merged
merged 2 commits into from
Jan 16, 2023
Merged

Link CleanRL's PPO Implementation #2780

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e90d678
Reference CleanRL's Efficient Implementation
vwxyzjn Jan 9, 2023
a621fbe
quick change
vwxyzjn Jan 9, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions examples/ppo/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,9 @@ tested on the following:
| --- | --- | --- | --- | --- |
| 1x V100 GPU | Breakout | 9h 15m 15s | 40M | [2020-10-02](https://tensorboard.dev/experiment/pY7D2qYQQLO9ZT5lA9PFPA) |

> **Note**
> It is possible to improve training efficiency through further optimizations. For example, CleanRL's PPO ([ppo_atari_envpool_xla_jax_scan.py](https://docs.cleanrl.dev/rl-algorithms/ppo/#ppo_atari_envpool_xla_jax_scanpy)) can achieve the same level of results in just 30 minutes with an RTX 2080 TI, 8 CPU, and the same hyperparameters — **a 1850% speedup end-to-end**. It achieves this by using [EnvPool](https://envpool.readthedocs.io/en/latest/), a library for fast parallelizing environments, jitting the entire rollout through [EnvPool's XLA interface](https://envpool.readthedocs.io/en/latest/content/xla_interface.html), storing data more efficiently, and `jax.scan`.

## How to run

Running `python ppo_main.py` will run the example with default
Expand Down