quangr

> 你好@quangr，感谢您的贡献！对于很多人来说，能够使用 JAX+PPO+MuJoCo+EnvPool 将改变游戏规则！此 PR 也将使#217变得不必要。 > > 一些评论和想法： > > * 您介意分享您的 wandb 用户名以便我将您添加到该`openrlbenchmark`实体吗？如果您可以在那里贡献跟踪实验，那就太好了，我们可以使用我们的 CLI 实用程序 ( https://github.com/openrlbenchmark/openrlbenchmark ) 来绘制图表。 Thank @51616 and @vwxyzjn for code reviewing[❤️](https://emojipedia.org/red-heart/)! My wandb...

I have submit new commits for most of the comments, and here are my answers to some other comments. If there is something missing, please help let me know >...

I have run experiment for Ant-v4 HalfCheetah-v4 Hopper-v4 Walker2d-v4 Swimmer-v4 Humanoid-v4 Reacher-v4 InvertedPendulum-v4 InvertedDoublePendulum-v4. Here is the report https://wandb.ai/openrlbenchmark/cleanrl/reports/MuJoCo-jax-EnvPool--VmlldzozNDczNDkz. # comparing the result to tianshou I notice that tianshou use...

> Thanks for running the results! They look great. Feel free to click `resolve conversation` as you resolve the PR comments and let me know when it's ready for another...

I have document the questions you brought up and I am now ready for the code review. I would be happy to hear your feedback.

WOW, what a detailed AMAZING replay! Thanks for your kindness. It really opened my mind and I appreciate it. I still need to write some code to fully understand it,...