Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,64 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
task_categories:
|
| 6 |
+
- text-classification
|
| 7 |
+
tags:
|
| 8 |
+
- clarus
|
| 9 |
+
- clinical
|
| 10 |
+
- quad-coupling
|
| 11 |
+
- population-shift
|
| 12 |
+
- protocol-deviation
|
| 13 |
+
- site-variance
|
| 14 |
+
- endpoint-fragility
|
| 15 |
+
pretty_name: Clinical Quad Population Shift Protocol Deviation Site Variance Endpoint Fragility v0.2
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
Clinical Quad Population Shift Protocol Deviation Site Variance Endpoint Fragility v0.2
|
| 19 |
+
|
| 20 |
+
What this dataset does
|
| 21 |
+
|
| 22 |
+
It tests whether a model can detect when clinical trial endpoints lose credibility under quad coupling.
|
| 23 |
+
|
| 24 |
+
Quad nodes
|
| 25 |
+
|
| 26 |
+
- population_shift
|
| 27 |
+
- protocol_deviation_rate
|
| 28 |
+
- site_variance_level
|
| 29 |
+
- endpoint_fragility
|
| 30 |
+
|
| 31 |
+
Labels
|
| 32 |
+
|
| 33 |
+
0 coherent
|
| 34 |
+
- Stable population
|
| 35 |
+
- Low deviations
|
| 36 |
+
- Low site variance
|
| 37 |
+
- Endpoint robust
|
| 38 |
+
|
| 39 |
+
1 tradeoff
|
| 40 |
+
- Some drift exists
|
| 41 |
+
- Endpoint still usable
|
| 42 |
+
- Risk is present but not terminal
|
| 43 |
+
|
| 44 |
+
2 collapse
|
| 45 |
+
- Endpoint credibility breaks
|
| 46 |
+
- Drift becomes structural
|
| 47 |
+
- Site or deviation pressure overwhelms interpretation
|
| 48 |
+
|
| 49 |
+
What changed in v0.2
|
| 50 |
+
|
| 51 |
+
- Version bumped to make scorer changes visible
|
| 52 |
+
- New scorer with validation and id-based matching
|
| 53 |
+
- Confusion and error sampling added
|
| 54 |
+
- Risk score and rule_pred diagnostics added
|
| 55 |
+
|
| 56 |
+
Files
|
| 57 |
+
|
| 58 |
+
data/train.csv
|
| 59 |
+
data/test.csv
|
| 60 |
+
scorer.py
|
| 61 |
+
|
| 62 |
+
Run scoring
|
| 63 |
+
|
| 64 |
+
python scorer.py --preds_csv predictions.csv --gold_csv data/test.csv
|