feat: publish CLI orchestrator#30
Merged
ziv-lazarov-nagish merged 3 commits intonagishfrom Apr 26, 2026
Merged
Conversation
ziv-lazarov-nagish
changed the base branch from
feat/publish-utils-and-card
to
feat/publish-hf-ops-and-eval
2 tasks
ziv-lazarov-nagish
changed the title
ziv-lazarov-nagish
force-pushed
the
feat/publish-cli
branch
from
9db2698 to
9763fbf
Compare
This was referenced Apr 20, 2026
ziv-lazarov-nagish
force-pushed
the
feat/publish-hf-ops-and-eval
branch
from
f4135e1 to
ec1c80f
Compare
ziv-lazarov-nagish
force-pushed
the
feat/publish-cli
branch
2 times, most recently
from
7944cde to
1329527
Compare
ziv-lazarov-nagish
added a commit
that referenced
this pull request
Apr 23, 2026
Rebase side-effect: after #31 renamed test_publish_cli.py → test_publish_hf_ops.py, git's rename detection merged #30's CLI-orchestrator additions into the HF-ops file. Restore the intended separation — TestPublishIntegration lives in test_publish_cli.py, TestCheckRegression and TestPromote stay in test_publish_hf_ops.py.
AmitMY
approved these changes
Apr 23, 2026
ziv-lazarov-nagish
force-pushed
the
feat/publish-hf-ops-and-eval
branch
from
4a5474a to
0c3cf67
Compare
Completes the publish pipeline on top of the pure-helpers layer: - `publish/utils.py`: append `_eval_single`, `run_evaluation`, `check_regression`, `promote`. - `publish/publish.py`: 8-step orchestrator — convert → find manifest → eval → regression check → save manifest → model card → upload to `weekly` branch → promote tag. - `publish/__init__.py`: re-export `publish` and `main`. - `datasets/common.py`: rename `_ensure_datasets_registered` → `ensure_datasets_registered` (now a public API so `run_evaluation` can call it without reaching into a private name). - `pyproject.toml`: add `[hf]` (huggingface_hub>=0.20.0) and `[publish]` (`[hf]` + `[train]`) optional-dependency groups. - `.env.example`: append HF section (`HF_TOKEN`, `HF_MODEL_REPO`, `HF_MODEL_REVISION`, `XDG_CACHE_HOME` for sparks cache). - `.gitignore`: add `dist/`, `wandb/`, `*.log`. Review-comment fixes shipped here: - #1: dropped unused `os`, `datetime`, `UTC` imports from `publish.py`. - #4: `_ensure_datasets_registered` → `ensure_datasets_registered` (public). - #5: `class EvalArgs: pass` → `argparse.Namespace(...)`. - #6: `--regression-threshold` default `0.02` → `0.005`. - #7: `promote()` now raises `ValueError` on unresolved revision instead of silently passing the ref string through to `create_tag`. 11 new tests in `test_publish_cli.py` cover: - `check_regression`: no_baseline (no tags / download failure), pass (within threshold), fail (beyond threshold). - `promote`: tag found, branch found, unresolved raises `ValueError`. - `publish()` integration with every HF + eval boundary mocked: skip-eval + no-promote, skip-eval + promote, eval + regression pass + promote, eval + regression fail (no promote).
Rebase side-effect: after #31 renamed test_publish_cli.py → test_publish_hf_ops.py, git's rename detection merged #30's CLI-orchestrator additions into the HF-ops file. Restore the intended separation — TestPublishIntegration lives in test_publish_cli.py, TestCheckRegression and TestPromote stay in test_publish_hf_ops.py.
ziv-lazarov-nagish
changed the base branch from
feat/publish-hf-ops-and-eval
to
nagish
ziv-lazarov-nagish
force-pushed
the
feat/publish-cli
branch
from
dd28274 to
4ca9439
Compare
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
2 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.
Depends on: #31 —
feat/publish-hf-ops-and-evalSummary
Fourth of a 5-PR stack — final layer of the #22 split (layer 5, PR #32, adds
--dry-runon top). Adds the CLI entry point — an 8-step orchestrator that stitches the helpers from the earlier layers into a runnable publish pipeline.publish/publish.py— 8-step orchestrator: convert → find manifest → eval → regression check → save manifest → model card → upload toweeklybranch → promote tag (on regression pass).publish/__init__.py— re-exportpublishandmain.pyproject.toml— add[publish]optional group ([hf]+[train])..gitignore— adddist/,wandb/,*.log.Review-comment fixes in this layer
os,datetime,UTCimports frompublish.py--regression-thresholddefault0.02→0.005(Fixes #4, #5, #7 live in PR #31.)
Tests
4 new
TestPublishIntegrationcases — every HF + eval boundary mocked:Stack
5-PR stack — layers 1–4 split #22, layer 5 adds
--dry-run:feat/safetensors-runtime→nagish(PR feat: safetensors runtime loading + HF-aware resolver #28)feat/publish-utils-and-card→ feat: safetensors runtime loading + HF-aware resolver #28 (PR feat: publish utils — conversion, versioning, model card #29)feat/publish-hf-ops-and-eval→ feat: publish utils — conversion, versioning, model card #29 (PR feat: publish HF ops + evaluation helpers #31)feat/publish-dry-run→ this PR (PR feat: --dry-run flag on publish CLI #32)When layers 1–4 merge, #22 can be closed.
Test plan
uv run --extra dev pytest sign_language_segmentation/tests— 103 passed (full stack, includes the 4 new integration tests)uv run --extra dev ruff check sign_language_segmentation/publish/— clean--dry-runagainstsign/segmentationon all datasets (see PR feat: --dry-run flag on publish CLI #32): all 8 steps ran — convert, manifest, eval (full), regression check, save manifest, model card render — with the final HF upload calls skipped. Model card inpublish_dry_run.mdmatches template. HF upload + tag promotion remain covered by the mocked integration tests above.