docs: training section #276
Draft
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
lbliii
force-pushed
the
llane/models-section
branch
from
061648f to
e448ab2
Compare
…uides Training section: - Add dataset format specification and validation guides - Add NeMo RL integration guides (SFT, DPO, GRPO, advanced usage) - Add rollout collection patterns and optimization guides - Add sampling strategies and performance tuning - Add data quality guidelines and resource server integration Setup-Deployment section: - Add configuration guides (debugging, multi-server, reference) - Add deployment guides (local, distributed computing, vLLM integration) - Add operations guides (testing, monitoring, profiling, debugging) Code changes: - Rename example_stateful_counter to stateful_counter - Remove mini_swe_agent (deprecated) - Add openai_judge_model config - Update resource server configs and scripts Signed-off-by: Lawrence Lane <[email protected]>
- Convert all {doc} links to {ref} targets in training and setup-deployment sections for better maintainability
- Fix reference target placement (move before headers for proper linking)
- Replace non-existent page links (evaluation, research, validation) with bold text
- Remove obsolete tutorial reference (custom-resource-server)
- Remove trailing horizontal rules from nemo-rl integration guides (Sphinx error)
- Fix get-started cross-references to use correct gs- prefix targets
Signed-off-by: Lawrence Lane <[email protected]>
…ries and fixing README link - Remove empty placeholder files (grpo.md, sft.md, dpo.md, troubleshooting.md) from nemo-rl toctree - Update docs/README.md to reference tutorials/index.md instead of non-existent tutorials/README.md - Reduces build warnings from 500+ to manageable level Signed-off-by: Lawrence Lane <[email protected]>
…tutorial - Remove non-existent 'multiple-environments' anchor link to how-to-faq - Update Configuration System reference to note it includes multi-environment setup - Resolves final [myst.xref_missing] warning Signed-off-by: Lawrence Lane <[email protected]>
- Create grpo.md for Group Relative Policy Optimization guide - Create sft.md for Supervised Fine-Tuning guide - Create dpo.md for Direct Preference Optimization guide - Create troubleshooting.md for integration troubleshooting - Add all four to toctree in nemo-rl/index.md - Each placeholder includes title, overview, and links to related resources - Resolves 'unknown document' warnings for nemo-rl integration guides Signed-off-by: Lawrence Lane <[email protected]>
…o match other index files Signed-off-by: Lawrence Lane <[email protected]>
…natural sections - Remove admonitions in favor of natural narrative flow - Add architectural explanation showing verify() lives in resource servers - Include concrete code examples from structured_outputs and mcqa servers - Add intro sentences to all section headers - Apply progressive disclosure from concept → architecture → examples → next steps - Make explicit that choosing resource server = choosing verification logic Signed-off-by: Lawrence Lane <[email protected]>
…MyST features - Convert to proper index/overview format with Topics section - Use dropdown for architecture details to reduce cognitive overload - Convert server comparison to list-table for scannability - Add verification workflow diagram and quick decision guide - Use list-table for actionable decision making - Add Related Topics section for workflow context - Keep resource server dependency explanation but in overview format Signed-off-by: Lawrence Lane <[email protected]>
Signed-off-by: Lawrence Lane <[email protected]>
lbliii
force-pushed
the
llane/docs-training
branch
from
dad24ef to
c39d151
Compare
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.
this is an exploration that will likely get scrapped