[RLlib] Add "official" benchmark script for Atari PPO benchmarks (new API stack). #45697
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: sven1977 <[email protected]>
Signed-off-by: sven1977 <[email protected]>
…nce_env_render_examples
Signed-off-by: sven1977 <[email protected]>
Signed-off-by: sven1977 <[email protected]>
Signed-off-by: sven1977 <[email protected]>
sven1977
requested review from
ArturNiederfahrenhorst and
simonsays1980
as code owners
Signed-off-by: sven1977 <[email protected]>
Signed-off-by: sven1977 <[email protected]>
simonsays1980
approved these changes
Jun 5, 2024
| commands.append(f"--env={env_name}") | ||
| commands.append(f"--wandb-run-name={env_name}") | ||
| print(f"Running {env_name} through command line=`{commands}`") | ||
| subprocess.run(commands) |
There was a problem hiding this comment.
It is somehow strange to me that we emulate running from the command line which in turn runs a script that could have been triggered direclty in the loop. It makes sense to me that users can run single envs, but why not triggering them directly in the loop?
| # AgileRL: https://github.com/AgileRL/AgileRL?tab=readme-ov-file#benchmarks | ||
| # [0] = reward to expect for DQN rainbow [1] = timesteps to run (always 200M for DQN | ||
| # rainbow). | ||
| # Note that for PPO, we simply run everything for 6M ts. |
There was a problem hiding this comment.
In atari_ppo.py the timesteps are set to 3M.
| parser = add_rllib_example_script_args() | ||
| parser = add_rllib_example_script_args( | ||
| default_reward=float("inf"), | ||
| default_timesteps=3000000, |
There was a problem hiding this comment.
Here we have set 3M timesteps while above in the benchmark_atari_ppo.py we comment on 6M.
richardsliu
pushed a commit
to richardsliu/ray
that referenced
this pull request
Jun 12, 2024
… API stack). (ray-project#45697) Signed-off-by: Richard Liu <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add "official" benchmark script for Atari PPO benchmarks (new API stack).
tuned_examplescript passing through some command line args.Why are these changes needed?
Related issue number
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.