refactor: use hf Dataset instead of pandas DataFrame in RLHFDataset #890
Conversation
This file contains hidden or 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
HF Dataset provides better memory management and can handle larger datasets. It also supports multi-process acceleration during map/filter operations (while pandas requires version >2.0). Now we can specify `filter_overlong_prompts` on large-scale datasets.
|
|
hiyouga
previously approved these changes
Apr 3, 2025
eric-haibin-lin
approved these changes
Apr 4, 2025
yushengsu-thu
pushed a commit
to yushengsu-thu/verl
that referenced
this pull request
Apr 4, 2025
…FDataset for speedup (volcengine#890) HF Dataset provides better memory management and can handle larger datasets. It also supports multi-process acceleration during map/filter operations (while pandas requires version >2.0). Now we can specify `filter_overlong_prompts` on large-scale datasets when set `filter_overlong_prompts_workers` to a appreciate num. --------- Co-authored-by: hoshi-hiyouga <[email protected]>
|
One of the breaking behavior change on this would be we can no longer point a directory as the |
Collaborator
Author
|
HI @twangnyc, I think you can use wildcard if you want to use the entire directory as your dataset. for example, you can set your |
|
@0x404 Thanks for your reply! Yes we are using wildcard to temporary fix our failing code running previously. Noticed you raised the new PR to make it compatible with pd.Dataframe behavior. Appreciate it! |
yuchenwang3
pushed a commit
to yuchenwang3/verl
that referenced
this pull request
Apr 25, 2025
…FDataset for speedup (volcengine#890) HF Dataset provides better memory management and can handle larger datasets. It also supports multi-process acceleration during map/filter operations (while pandas requires version >2.0). Now we can specify `filter_overlong_prompts` on large-scale datasets when set `filter_overlong_prompts_workers` to a appreciate num. --------- Co-authored-by: hoshi-hiyouga <[email protected]>
histmeisah
pushed a commit
to SJTU-IAAR/verl
that referenced
this pull request
Apr 27, 2025
…FDataset for speedup (volcengine#890) HF Dataset provides better memory management and can handle larger datasets. It also supports multi-process acceleration during map/filter operations (while pandas requires version >2.0). Now we can specify `filter_overlong_prompts` on large-scale datasets when set `filter_overlong_prompts_workers` to a appreciate num. --------- Co-authored-by: hoshi-hiyouga <[email protected]>
6 tasks
vermouth1992
pushed a commit
that referenced
this pull request
May 7, 2025
### Checklist Before Starting - [x] Search for similar PR(s). ### What does this PR do? This PR update some outdated docs on config: - Add `filter_overlong_prompts_workers` configuration option, which introduced in #890 - Add documentation for `actor_rollout_ref.rollout.val_kwargs` parameters, fix #1352 - Fix attribution of several configuration options to their proper namespaces ### High-Level Design > Demonstrate the high-level design if this PR is complex. ### Specific Changes > List the specific changes. ### API > Demonstrate how the API changes if any. ### Usage Example > Provide usage example(s) for easier usage. ```python # Add code snippet or script demonstrating how to use this ``` ### Test > For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluatuion results, etc. ### Additional Info. - **Issue Number**: Fixes issue # or discussion # if any. - **Training**: [Note which backend this PR will affect: FSDP, Megatron, both, or none] - **Inference**: [Note which backend this PR will affect: vLLM, SGLang, both, or none] ### Checklist Before Submitting - [x] Read the [Contribute Guide](https://github.com/volcengine/verl?tab=readme-ov-file#contribution-guide). - [x] Apply [pre-commit checks](https://github.com/volcengine/verl?tab=readme-ov-file#code-linting-and-formatting). - [x] Add `[BREAKING]` to the PR title if it breaks any API. - [x] Update the documentation about your changes in the [docs](https://github.com/volcengine/verl/tree/main/docs). - [x] Add CI test(s) if neccessary.
chenjiaoAngel
added a commit
to chenjiaoAngel/verl
that referenced
this pull request
Nov 14, 2025
…FDataset for speedup (volcengine#890) HF Dataset provides better memory management and can handle larger datasets. It also supports multi-process acceleration during map/filter operations (while pandas requires version >2.0). Now we can specify `filter_overlong_prompts` on large-scale datasets when set `filter_overlong_prompts_workers` to a appreciate num. --------- Co-authored-by: hoshi-hiyouga <[email protected]>
chenjiaoAngel
added a commit
to chenjiaoAngel/verl
that referenced
this pull request
Nov 14, 2025
…ine#1355) ### Checklist Before Starting - [x] Search for similar PR(s). ### What does this PR do? This PR update some outdated docs on config: - Add `filter_overlong_prompts_workers` configuration option, which introduced in volcengine#890 - Add documentation for `actor_rollout_ref.rollout.val_kwargs` parameters, fix volcengine#1352 - Fix attribution of several configuration options to their proper namespaces ### High-Level Design > Demonstrate the high-level design if this PR is complex. ### Specific Changes > List the specific changes. ### API > Demonstrate how the API changes if any. ### Usage Example > Provide usage example(s) for easier usage. ```python # Add code snippet or script demonstrating how to use this ``` ### Test > For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluatuion results, etc. ### Additional Info. - **Issue Number**: Fixes issue # or discussion # if any. - **Training**: [Note which backend this PR will affect: FSDP, Megatron, both, or none] - **Inference**: [Note which backend this PR will affect: vLLM, SGLang, both, or none] ### Checklist Before Submitting - [x] Read the [Contribute Guide](https://github.com/volcengine/verl?tab=readme-ov-file#contribution-guide). - [x] Apply [pre-commit checks](https://github.com/volcengine/verl?tab=readme-ov-file#code-linting-and-formatting). - [x] Add `[BREAKING]` to the PR title if it breaks any API. - [x] Update the documentation about your changes in the [docs](https://github.com/volcengine/verl/tree/main/docs). - [x] Add CI test(s) if neccessary.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
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.
HF Dataset provides better memory management and can handle larger datasets. It also supports multi-process acceleration during map/filter operations (while pandas requires version >2.0).
Now we can specify
filter_overlong_promptson large-scale datasets when setfilter_overlong_prompts_workersto a appreciate num.