feat: add release verification test suite - #56
Merged
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
Adds an opt-in, real-data “release verification” test suite intended for manual post-release validation of end-to-end training/infer/eval across classification, localization, and VQA model families, plus small supporting runtime improvements.
Changes:
- Introduces
tests/release/with shared helpers, detailed documentation, and task-specific release verification tests gated byRUN_OSL_RELEASE_TESTS=1. - Adds a convenience runner script for the release suite and ignores the generated cache/output directory.
- Extends core training support by adding SGD optimizer construction and aligning ResNet feature extractors with the shared
BaseExtractFeaturesbase class.
Reviewed changes
Copilot reviewed 8 out of 10 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
| tests/release/test_vqa_release.py | New VQA release verification tests (LoRA backends) + sharded dataset fixture |
| tests/release/test_localization_release.py | New localization release verification tests (E2E + feature-based families) |
| tests/release/test_classification_release.py | New classification release verification tests across backbone families |
| tests/release/README.md | Documents release test goals, datasets, prerequisites, and tuning knobs |
| tests/release/_release_common.py | Shared gating, HF download helpers, config materialization, and run output paths |
| tests/release/init.py | Package init enabling relative imports within tests/release/ |
| scripts/run_release_tests.sh | Guarded wrapper to run the release suite manually |
| opensportslib/models/backbones/builder.py | ResNet extractor now derives from shared BaseExtractFeatures |
| opensportslib/core/optimizer/builder.py | Adds SGD support (and explicit error on unknown optimizer type) |
| .gitignore | Ignores .release_test_cache/ generated by release verification runs |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
| assert pred_path.exists() | ||
|
|
||
| report_step(f"[{run_name}] evaluate()") | ||
| metrics = model.evaluate(test_set=str(split_paths["test"]), use_wandb=False) |
| assert pred_path.exists() | ||
|
|
||
| report_step(f"[{run_name}] evaluate()") | ||
| metrics = model.evaluate(test_set=str(split_paths["test"]), use_wandb=False) |
| download_files, | ||
| download_shard_split, | ||
| epochs_for, | ||
| hf_token, |
| valid_set=str(split_paths["valid"]), | ||
| use_wandb=False, | ||
| ) | ||
| assert checkpoint, f"[{run_name}] train() did not return a checkpoint" |
| # invoked directly, and requires RUN_OSL_RELEASE_TESTS=1 to actually execute | ||
| # anything (the tests skip themselves otherwise). | ||
|
|
||
| set -uo pipefail |
Collaborator
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
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.
Summary\n- add opt-in release verification tests for classification, localization, and VQA\n- add shared dataset/config/cache helpers and a guarded runner script\n- add SGD optimizer support and align ResNet extractors with the shared base class\n- ignore generated release-test cache output\n\n## Testing\n- passed\n- and could not run because pytest is not installed in the active environment