Add cellpose v4 support; migrate to npe2 and pyproject.toml#62
Open
jeskowagner wants to merge 9 commits into
Open
Add cellpose v4 support; migrate to npe2 and pyproject.toml#62jeskowagner wants to merge 9 commits into
jeskowagner wants to merge 9 commits into
Conversation
The plugin now detects the installed cellpose major version at import time and adapts accordingly. On v4, model/channel selection and image- based diameter estimation are hidden because cellpose-sam only provides the one model. Other changes: - Migrate from npe1 (napari-plugin-engine) to npe2: remove legacy hook registrations, add napari.yaml manifest with valid python_name entries - Replace setup.cfg/requirements.txt/tox.ini/MANIFEST.in with a single pyproject.toml (tox config embedded under [tool.tox]) - Remove the Cells 2D sample data contribution; the source URL was broken - Guard _compute_diameter_shape against no shape layer being selected
carsen-stringer
self-requested a review
April 20, 2026 17:16
Author
|
Just saw the unit test failures, will investigate in the next days and ping again when fixed :) |
- For cellpose V4, show z_axis and channel_axis dropdowns for 4D images (required)
and 3D images (optional; -1 sentinel = pure spatial ZYX); defaults inferred
from napari axis labels
- Fix 3D flow/cellprob handling: skip resize_image for 3D outputs to avoid
collapsing the Z axis and leaving layers in a broken vispy state
- Remove output_flows/output_outlines checkboxes; always write all output layers
- Replace sample data source with OSF zip (rgb_2D_tif + rgb_3D); add
rgb_2D sample ("Cells (2D+2Ch)") to napari.yaml
- Exclude build/ directory from setuptools package discovery
- Update tests: switch sample from local file to 'rgb_2D' napari sample, fix
dock_widgets API (drop underscore prefix), update expected mask counts for
cellpose V4 vs V3, and adjust diameter tolerance for Windows/CPU
- Rewrite sample data loader: fetch from OSF zip instead of per-file URLs,
add rgb_2D sample, set axis_labels on rgb_3D so z/channel dropdowns
auto-populate correctly
Author
|
@carsen-stringer I have changed out some of the tests - Cellpose-SAM on CPU in 3D simply takes too long. I also had to make sure that torch is loaded in before pyqt due to pytorch/pytorch#166628. I added some new tests, and a CI test for cellpose=3 to make sure the GUI stays compatible. With these changes in place, all tests are passing now! |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #62 +/- ##
==========================================
- Coverage 54.31% 52.98% -1.33%
==========================================
Files 3 3
Lines 232 285 +53
==========================================
+ Hits 126 151 +25
- Misses 106 134 +28 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Author
|
Looks like there were two problems with CI:
|
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.
The plugin now detects the installed cellpose major version at import time and adapts accordingly. On v4, model/channel selection and image- based diameter estimation are hidden because cellpose-sam only provides the one model.
Other changes:
Fixes #61.