Instructions to use adarshcod30/margadrishti-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Scikit-learn
How to use adarshcod30/margadrishti-models with Scikit-learn:
from huggingface_hub import hf_hub_download import joblib model = joblib.load( hf_hub_download("adarshcod30/margadrishti-models", "sklearn_model.joblib") ) # only load pickle files from sources you trust # read more about it here https://skops.readthedocs.io/en/stable/persistence.html - Notebooks
- Google Colab
- Kaggle
File size: 1,725 Bytes
55151a3 dc392dc 55151a3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | # Model Card · B3 — Is this event High or Low priority?
*Generated 2026-09-02 from `reports/` — do not edit by hand.*
## What this model answers
Is this event High or Low priority?
## Caveats
- NOT A PREDICTION (F13). `priority` is a deterministic operational rule: corridor status predicts High at 99.84%, with 13 exceptions in 8,173 events.
- Scores near 1.0 reflect that rule, not learned structure. Reporting them as modelling performance would be misleading.
## Results
Ranked by **PR-AUC ↑** (higher is better). Every family that ran is listed, including those that lost — the comparison is the deliverable. Figures are means over 5 chronological folds, ± one standard deviation.
| Model | Family | PR-AUC ↑ | ROC-AUC ↑ | Brier ↓ | Best F1 ↑ | Fit | Beats baseline |
|---|---|---:|---:|---:|---:|---:|---|
| `catboost` | gbdt | 0.9999 ± 0.000 | 0.9998 | 0.0022 | 0.9990 | 2s | yes |
| `hist_gbm` | gbdt | 0.9998 ± 0.000 | 0.9997 | 0.0032 | 0.9985 | 2s | yes |
| `lightgbm` | gbdt | 0.9997 ± 0.001 | 0.9997 | 0.0016 | 0.9992 | 3s | yes |
| `xgboost` | gbdt | 0.9994 ± 0.001 | 0.9992 | 0.0028 | 0.9982 | 1s | yes |
| `random_forest` | trees | 0.9974 ± 0.005 | 0.9980 | 0.0394 | 0.9975 | 0s | yes |
| `extra_trees` | trees | 0.9972 ± 0.005 | 0.9977 | 0.0730 | 0.9975 | 0s | yes |
| `logistic` | linear | 0.9934 ± 0.008 | 0.9951 | 0.0060 | 0.9975 | 0s | yes |
| `prior` | baseline | 0.6146 ± 0.024 | 0.5000 | 0.2370 | 0.7611 | 0s | *baseline* |
**Best learned family: `catboost`** at 0.9999, +62.7% against the strongest baseline (`prior` at 0.6146).
## Reproduction
```bash
make data && make ingest && make features
make train-b
```
Seed `42`. Splits are chronological, never random.
|