docs: expand dataset formats into a full reference guide - #465
Open
O7c2 wants to merge 3 commits into
Open
Conversation
added 3 commits
August 11, 2026 15:33
Add step-by-step cookbook tutorial covering custom dataset loader and reward function authoring, with worked examples, debugging tips, and cross-references from concept guides.
State that the current runtime is AReno-owned CUDA/NVIDIA code, and explicitly list MLX, TPU, vLLM, SGLang, Megatron, and Ray as out of scope per issue inclusionAI#58. Closes inclusionAI#58
Rewrite the Dataset Formats page with mental model, concrete schema examples per algorithm (SFT with pre-tokenized rows, RL with GSM8K walkthrough, DPO with both prompt/response and chat message pairs, agentic), working CLI commands, loader contract, and schema summary table. State plainly that raw GSM8K is neither RL-ready nor SFT-ready. Closes inclusionAI#70
Contributor
Author
1 similar comment
Contributor
Author
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do?
Rewrite the Dataset Formats concept page as a comprehensive reference guide covering the mental model (raw data →
loader → schema), all training schemas with concrete examples and working CLI commands, GSM8K walkthrough, and a
schema summary table.
Related issue
See #70
Type of change
How was it tested?
Built docs with
sphinx-build -b html -W . _build/html— zero warnings.Checklist
pytest tests/ -k cpu). (skipped — doc-only change)