Add rnd_ppo.py documentation and refactor

[yooceii]

Description

Closes #127

Types of changes

• [ ] Bug fix
• [ ] New feature
• [ ] New algorithm
• [x] Documentation

Checklist:

• [x] I've read the CONTRIBUTION guide (required).
• [x] I have ensured pre-commit run --all-files passes (required).
• [x] I have updated the documentation and previewed the changes via mkdocs serve.
• [ ] I have updated the tests accordingly (if applicable).

If you are adding new algorithms or your change could result in performance difference, you may need to (re-)run tracked experiments. See https://github.com/vwxyzjn/cleanrl/pull/137 as an example PR.

• [x] I have contacted @vwxyzjn to obtain access to the openrlbenchmark W&B team (required).
• [ ] I have tracked applicable experiments in openrlbenchmark/cleanrl with --capture-video flag toggled on (required).
• [ ] I have added additional documentation and previewed the changes via mkdocs serve.
  • [x] I have explained note-worthy implementation details.
  • [ ] I have explained the logged metrics.
  • [x] I have added links to the original paper and related papers (if applicable).
  • [ ] I have added links to the PR related to the algorithm.
  • [ ] I have created a table comparing my results against those from reputable sources (i.e., the original paper or other reference implementation).
  • [ ] I have added the learning curves (in PNG format with width=500 and height=300).
  • [ ] I have added links to the tracked experiments.
• [ ] I have updated the tests accordingly (if applicable).

[vercel[bot]]

This pull request is being automatically deployed with Vercel (learn more).
To see the status of your deployment, click below or on the icon next to each commit.

🔍 Inspect: https://vercel.com/vwxyzjn/cleanrl/FVfp6xKi7pTtnaXPFTa7dhPRKnqL
✅ Preview: https://cleanrl-git-rnd-doc-vwxyzjn.vercel.app

[gitpod-io[bot]]

[vercel[bot]]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Updated
cleanrl	✅ Ready (Inspect)	Visit Preview	Aug 25, 2022 at 3:56AM (UTC)

[vwxyzjn]

Hey, @yooceii would you mind reverting the formatting change? They make it harder to review and identify the code specifically relating to RND.

Formatting change should be done in a larger PR in #167 together...

[vwxyzjn]

Refactor Check (compatible with the performance in the tracked experiment)

I compared the SPS performance of the latest refactor against the old script used in the tracked experiment. I named the old one old_ppo_rnd_envpool.py and ran it and the latest script without initializing observation normalization parameter.

The following screenshot confirms the refactor does not result in a performance difference (and that we did the refactor correctly — making it faster without impacting sample efficiency)