feat(protpardelle): add pixi env and bake into image/CI by xraymemory · Pull Request #331 · diff-use/sampleworks

xraymemory · 2026-07-21T18:58:25Z

Summary

Splits the Protpardelle environment + image plumbing out of #300 so the runtime image and CI can carry Protpardelle before the wrapper/correctness work lands. Once this merges to main, the pixi-with-checkpoints image rebuilds with the env baked in, and #300 can rebase to just the wrapper code + tests.

Add the [tool.pixi.feature.protpardelle] feature and protpardelle / protpardelle-dev environments (lifted from feat(add protpardelle): Add the Protpardelle-1c cc89 protein generative model. #300, git dep pinned to ee378400… — the commit the wrapper targets — for reproducible builds).
Install the protpardelle env in the Docker pixi-envs layer alongside boltz/protenix/rf3/analysis.
Add protpardelle-dev to the typecheck and cpu-tests CI matrices.
Regenerate pixi.lock.

On the chicken/egg

There isn't a hard one. CI installs each pixi env fresh via setup-pixi from this branch's pyproject.toml + pixi.lock — it does not consume the pre-built image — so this PR's own CI exercises protpardelle-dev immediately. The Docker image is only needed for cluster/ACTL sampling and rebuilds from main on merge; the env definition and Dockerfile change travel together here, so nothing depends on the image pre-existing. The split is for de-risking (validate the heavy git/CUDA env solves and the image builds independently) and to get the image published ahead of the correctness PR — not because of a build-order dependency.

Note on tests / checkpoints

Provisioning weights is not required for the Protpardelle tests: tests/models/protpardelle/conftest.py points PROTPARDELLE_MODEL_PARAMS at a throwaway dir and builds a small randomly-initialized CPU model, and importorskips if the package is absent. So installing the env is sufficient to un-skip them in CI. Baking real checkpoints + PROTPARDELLE_MODEL_PARAMS into the image is a follow-up needed only for actual sampling / GPU integration.

Follow-ups (not in this PR)

Verify pytorch resolves to a CUDA build in the linux-64 env; if it lands the CPU wheel, move it to torch under pypi-dependencies as flagged in feat(add protpardelle): Add the Protpardelle-1c cc89 protein generative model. #300 review.
Bake Protpardelle checkpoints + model-params for real sampling runs.

Validation

pixi lock regenerated cleanly.
CI on this branch runs protpardelle-dev typecheck + cpu-tests.

Summary by CodeRabbit

New Features
- Added Protpardelle environments for development and runtime workflows.
- Added support for installing Protpardelle and its required dependencies in Docker images.
Tests
- Expanded type-checking and CPU test coverage to include the new Protpardelle development environments.

Add the protpardelle feature and protpardelle/protpardelle-dev environments (git dep pinned to ee378400 for reproducibility), install the env in the Docker pixi-envs layer, and add protpardelle-dev to the typecheck and cpu-tests CI matrices. Splits the environment/image plumbing out of #300 so the runtime image and CI carry Protpardelle before the wrapper work lands.

coderabbitai · 2026-07-21T18:58:44Z

Warning

Review limit reached

@xraymemory, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 16 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: c3db2e47-93fd-427d-b7d8-fdc38b0a3859

📥 Commits

Reviewing files that changed from the base of the PR and between 2642556 and 3afa8ee.

⛔ Files ignored due to path filters (1)

pixi.lock is excluded by !**/*.lock

📒 Files selected for processing (2)

.github/workflows/ci.yml
pyproject.toml

📝 Walkthrough

Walkthrough

Adds Protpardelle Pixi dependencies and environments, installs the environment in the Docker image, and includes protpardelle-dev in typecheck and CPU-test CI matrices.

Changes

Protpardelle environment integration

Layer / File(s)	Summary
Define Protpardelle environments `pyproject.toml`	Adds the Protpardelle feature, dependency pins, platform support, and `protpardelle`/`protpardelle-dev` environments.
Integrate build and CI coverage `Dockerfile`, `.github/workflows/ci.yml`	Installs the Protpardelle environment in Docker and adds its development environment to typecheck and CPU-test matrices.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

diff-use/sampleworks#189: Updates the GitHub Actions environment matrices, including protenix-dev.

Suggested reviewers: mag-astera

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: adding the Protpardelle Pixi environment and wiring it into Docker and CI.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch mdc/protpardelle-env-image

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

Copilot

Pull request overview

This PR prepares the codebase/CI/runtime image to support the upcoming Protpardelle wrapper work by adding a dedicated Pixi feature + environments and wiring them into the Docker build and GitHub Actions matrices, with a regenerated lockfile for reproducible solves. (Confidence: ~85%)

Changes:

Add a new [tool.pixi.feature.protpardelle] feature with pinned git dependency (ee378400…) and define protpardelle / protpardelle-dev environments in pyproject.toml.
Bake the protpardelle environment into the pixi-envs Docker build layer alongside existing model envs.
Expand CI matrices to run ty check and CPU tests under protpardelle-dev as well (and regenerate pixi.lock accordingly).

Reviewed changes

Copilot reviewed 3 out of 4 changed files in this pull request and generated 2 comments.

File	Description
pyproject.toml	Adds the Protpardelle Pixi feature + environments (and associated dependency pins).
Dockerfile	Installs the new `protpardelle` Pixi environment in the image’s consolidated env layer.
.github/workflows/ci.yml	Adds `protpardelle-dev` to the typecheck + CPU test matrices.
pixi.lock	Regenerated to lock the new feature/environments for `--frozen` installs.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

marcuscollins · 2026-07-21T19:21:48Z

+[tool.pixi.feature.protpardelle.pypi-options.dependency-overrides]
+pandas = "==2.3.1"


This can be removed.

marcuscollins · 2026-07-21T19:22:41Z

+protpardelle = {features = ["protpardelle"]}
+protpardelle-dev = {features = ["protpardelle", "dev"]}


Probably should update those tasks as well.

The shared test_at_least_one_available asserts one of boltz/protenix/rf3 is importable, which fails in a protpardelle-only env. main has no protpardelle tests or availability flag yet, so the cpu-tests matrix entry adds no coverage here; it moves to #300 with the wrapper and the PROTPARDELLE_AVAILABLE update. Keep protpardelle-dev in typecheck and in the image.

xraymemory · 2026-07-21T19:12:40Z

Pushed a fix. First run failed only on tests (protpardelle-dev): the shared tests/test_imports.py::test_at_least_one_available asserts one of BOLTZ/PROTENIX/RF3_AVAILABLE, which is false in a protpardelle-only env. main has no Protpardelle wrapper, availability flag, or tests yet (those are in #300), so running the generic suite under protpardelle-dev adds no coverage and trips that cross-model assertion.

Dropped protpardelle-dev from the cpu-tests matrix here; it moves to #300 alongside the PROTPARDELLE_AVAILABLE update and the actual wrapper tests. Kept it in typecheck (passing) and in the Docker pixi-envs layer, which are the infra deliverables.

marcuscollins

A couple small changes would improve this, but it is fine as-is, so I'll let you fix and merge without re-reviewing.

marcuscollins · 2026-07-21T19:20:48Z

+typer = "*"
+
+[tool.pixi.feature.protpardelle.pypi-dependencies]
+protpardelle = {git = "https://github.com/ProteinDesignLab/protpardelle-1c.git", rev = "ee378400f25b801fa481028000f9060183d7fb4c"}


the various code review agents have told me that we should put the torch requirement in here as torch = ">=2.6.0,<2.8"

marcuscollins · 2026-07-21T19:21:48Z

+[tool.pixi.feature.protpardelle.pypi-options.dependency-overrides]
+pandas = "==2.3.1"


This can be removed.

marcuscollins · 2026-07-21T19:22:41Z

+protpardelle = {features = ["protpardelle"]}
+protpardelle-dev = {features = ["protpardelle", "dev"]}


Probably should update those tasks as well.

…wire tasks - Move torch to pypi-dependencies as torch>=2.6.0,<2.8 so linux-64 gets the CUDA wheel instead of conda's cpu_generic build. - Remove the per-feature pandas override that duplicated the workspace-level one. - Add protpardelle-dev to the test-all/test-fast cross-env tasks per the 'new model env needs entries' note.

xraymemory · 2026-07-21T19:42:21Z

Addressed all three review points (pushed in 3afa8ee):

torch → pypi — moved to torch = ">=2.6.0,<2.8" under pypi-dependencies. Confirmed via the regenerated lock: protpardelle-dev drops the conda pytorch … cpu_generic build and now pulls the pypi wheel with nvidia-*-cu12 on linux-64, i.e. the CUDA build. Good catch — that was the real fix.
Duplicate pandas override — removed; the workspace-level [tool.pixi.pypi-options.dependency-overrides] is now the sole source.
Cross-env tasks — added protpardelle-dev to test-all and test-fast.

One heads-up on #3: until #300 lands the PROTPARDELLE_AVAILABLE flag, the protpardelle-dev leg of test-all/test-fast will fail test_at_least_one_available (same cross-model assertion that made me keep it out of the CI cpu-tests matrix). The tasks already continue-past-failure, and it self-heals when #300 merges — but that's why CI cpu-tests still excludes it while the local tasks now list it. Happy to defer the task entry to #300 instead if you'd rather keep pixi run test-all green in the interim; your call.

xraymemory requested review from a team, Abdelsalam-Abbas, denis-zaitsev and mag-astera as code owners July 21, 2026 18:58

Copilot AI review requested due to automatic review settings July 21, 2026 18:58

xraymemory had a problem deploying to gpu-testing July 21, 2026 18:58 — with GitHub Actions Error

Copilot started reviewing on behalf of xraymemory July 21, 2026 18:58 View session

Copilot AI reviewed Jul 21, 2026

View reviewed changes

xraymemory had a problem deploying to gpu-testing July 21, 2026 19:12 — with GitHub Actions Error

marcuscollins approved these changes Jul 21, 2026

View reviewed changes

xraymemory temporarily deployed to gpu-testing July 21, 2026 19:42 — with GitHub Actions Inactive

marcuscollins merged commit 7b6ab4e into main Jul 21, 2026
12 checks passed

coderabbitai Bot mentioned this pull request Jul 21, 2026

feat(add protpardelle): Add the Protpardelle-1c cc89 protein generative model. #300

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat(protpardelle): add pixi env and bake into image/CI#331

feat(protpardelle): add pixi env and bake into image/CI#331
marcuscollins merged 3 commits into
mainfrom
mdc/protpardelle-env-image

xraymemory commented Jul 21, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

coderabbitai Bot commented Jul 21, 2026 •

edited

Loading

Review limit reached

Walkthrough

Changes

Uh oh!

Copilot AI left a comment

Uh oh!

marcuscollins Jul 21, 2026

Uh oh!

marcuscollins Jul 21, 2026

Uh oh!

xraymemory commented Jul 21, 2026

Uh oh!

marcuscollins left a comment

Uh oh!

marcuscollins Jul 21, 2026

Uh oh!

marcuscollins Jul 21, 2026

Uh oh!

marcuscollins Jul 21, 2026

Uh oh!

xraymemory commented Jul 21, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		[tool.pixi.feature.protpardelle.pypi-options.dependency-overrides]
		pandas = "==2.3.1"

		protpardelle = {features = ["protpardelle"]}
		protpardelle-dev = {features = ["protpardelle", "dev"]}

Uh oh!

Conversation

xraymemory commented Jul 21, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

On the chicken/egg

Note on tests / checkpoints

Follow-ups (not in this PR)

Validation

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Jul 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review limit reached

Walkthrough

Changes

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

marcuscollins Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

marcuscollins Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

xraymemory commented Jul 21, 2026

Uh oh!

marcuscollins left a comment

Choose a reason for hiding this comment

Uh oh!

marcuscollins Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

marcuscollins Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

marcuscollins Jul 21, 2026

Choose a reason for hiding this comment

Uh oh!

xraymemory commented Jul 21, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

xraymemory commented Jul 21, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Jul 21, 2026 •

edited

Loading