Skip to content

docs: clarify resources server example-data contract and license validation#1560

Open
ananthsub wants to merge 2 commits into
NVIDIA-NeMo:mainfrom
ananthsub:ananthsub/docs-resources-server-checks
Open

docs: clarify resources server example-data contract and license validation#1560
ananthsub wants to merge 2 commits into
NVIDIA-NeMo:mainfrom
ananthsub:ananthsub/docs-resources-server-checks

Conversation

@ananthsub

Copy link
Copy Markdown
Contributor

Follow up for #1559

Document that a resources server's example data must include example.jsonl, example_metrics.json (from example_validation), and example_rollouts.jsonl, and that example.jsonl must be in materialized Responses format.

Also generalize license guidance to point at the validated enum in config_types.py so docs are in sync with the code

…dation

Document that a resources server's example data must include example.jsonl,
example_metrics.json (from example_validation), and example_rollouts.jsonl, and
that example.jsonl must be in materialized Responses format (example datasets are
not templated via prompt_config). Generalize license guidance to point at the
validated enum in config_types.py instead of an inline (and partly inaccurate)
list.

Signed-off-by: Ananth Subramaniam <ansubramania@nvidia.com>
**Data conversion**: Write conversion scripts in the **source repo** (e.g. your dataset repository), not in NeMo-Gym. Prompt files also belong in the source repo. Exception: when there is no external source repo. See `references/patterns.md` § "Data Conversion Script Pattern".

**`example.jsonl`**: Generate 5 entries for smoke testing. This file is committed directly to git in `data/example.jsonl`.
**`example.jsonl`**: Generate 5 entries for smoke testing, committed directly to git in `data/example.jsonl`. Each row must be in materialized Responses format (contain `responses_create_params`) — example datasets are not templated via `prompt_config`, so raw rows won't validate.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is this the same as generated materialized_inputs files?

@github-actions

Copy link
Copy Markdown

🌿 Preview your docs: https://nvidia-preview-ananthsub-docs-resources-server-checks.docs.buildwithfern.com/nemo/gym

Here are the markdown pages you've updated:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants