| Dimension | Shared | Only in A | Only in B | Overlap |
|---|---|---|---|---|
| Sections | 0 | 1 | 1 | 0% |
| Commands | 0 | 0 | 3 | 0% |
| Section tags | 0 | 2 | 3 | 0% |
What each file covers
Sections
0 shared · 1 only in A · 1 only in B- − ML Evaluation Memory
- + AI Service Rules
Commands
0 shared · 0 only in A · 3 only in B- + python -m mypy app
- + python -m ruff check .
- + python -m pytest
Section tags
0 shared · 2 only in A · 3 only in B- − code-style
- − performance
- + test
- + lint-format
- + do-not
Line diff
Aledon8/OpenLeukemia · ml/evaluation/CLAUDE.md
@@ −1 @@
1# ML Evaluation Memory
2
3Apply this inside `ml/evaluation/`.
4
5- Report validation setup, data splits, metrics, and known limitations.
6- Prefer clinically cautious interpretation of model quality.
7- Track false positives, false negatives, calibration, and subgroup concerns when relevant.
8- Evaluation artifacts should support review, not overstate readiness.
9- Summaries should state what the model can and cannot support.
10
Aledon8/OpenLeukemia · .cursor/rules/ai-service.mdc
@@ +1 @@
1---
2description: Python FastAPI AI-service rules
3globs:
4 - "ai-service/**/*.py"
5alwaysApply: false
6---
7
8# AI Service Rules
9
10- Keep FastAPI route handlers thin; put reusable behavior in `app/domains`.
11- Use explicit Pydantic request and response models.
12- Preserve strict typing and keep `python -m mypy app` clean.
13- Use `response_model` for route outputs.
14- Treat extraction and model outputs as explainable signals or candidate data, not clinical decisions.
15- Run `python -m ruff check .` and relevant `python -m pytest` tests for Python changes.
16
@@ −1 +1 @@
1−# ML Evaluation Memory
1+---
2+description: Python FastAPI AI-service rules
3+globs:
4+ - "ai-service/**/*.py"
5+alwaysApply: false
6+---
27
3−Apply this inside `ml/evaluation/`.
8+# AI Service Rules
49
5−- Report validation setup, data splits, metrics, and known limitations.
6−- Prefer clinically cautious interpretation of model quality.
7−- Track false positives, false negatives, calibration, and subgroup concerns when relevant.
8−- Evaluation artifacts should support review, not overstate readiness.
9−- Summaries should state what the model can and cannot support.
10+- Keep FastAPI route handlers thin; put reusable behavior in `app/domains`.
11+- Use explicit Pydantic request and response models.
12+- Preserve strict typing and keep `python -m mypy app` clean.
13+- Use `response_model` for route outputs.
14+- Treat extraction and model outputs as explainable signals or candidate data, not clinical decisions.
15+- Run `python -m ruff check .` and relevant `python -m pytest` tests for Python changes.
1016
