Repository navigation
docs: tile at each model's own scale for PathProfilerQC and Prism - #4
Merged
Merged
Conversation
- preprocessing, visualization: PathProfiler's QC model was trained on 256 px tiles at 5x (about 2 µm/px; upstream quality-assessment/run.py downsamples to 5X), not 0.5 mpp. Tile for it separately at mpp=2 and plot those tiles. - zero-shot-learning: Prism reads Virchow features of 224 px tiles at 0.5 mpp, not the 512 px tiles used for CONCH. Tile for it separately. Edited cells have their outputs cleared; the notebook workflow re-renders them. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
3 tasks done
Contributor
There was a problem hiding this comment.
Copilot review overview
🟢 Approval recommended
The Prism workflow consistently uses the dedicated tile key and correctly derived feature key.
Review effort: Balanced
Findings: None
What changed in this PR
Corrects tutorial preprocessing so PathProfilerQC and Prism use their model-specific tile scales.
Changes:
- Adds dedicated 224 px, 0.5 mpp Prism tiling.
- Uses Prism-specific Virchow feature keys.
- Clears affected notebook outputs.
| File | Description |
|---|---|
tutorials/zero-shot-learning.ipynb |
Uses correctly scaled Prism tiles and features. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
# Conflicts: # tutorials/preprocessing.ipynb # tutorials/visualization.ipynb # tutorials/zero-shot-learning.ipynb
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Tutorial fixes from the preprocessing audit of lazyslide-models (rendeirolab/lazyslide-models#31):
quality-assessment/run.pydownsamples to 5X before predicting. The tutorials ran it on the 256 px tiles at 0.5 mpp. They now tile separately atmpp=2intoqc_tilesand plot those tiles.prism_tiles. CONCH keeps its 512 px tiles.Outputs of the edited cells are cleared. The notebook workflow re-renders them.
Test plan
tile_tissues(wsi, 256, mpp=2)gives 13 tiles, andtile_predictionplus all threepl.tilescalls run