Version sync: sklearn 1.7.2 retrain + provenance
Browse files
README.md
CHANGED
|
@@ -30,7 +30,7 @@ model-index:
|
|
| 30 |
|
| 31 |
Predicts probability that a hotel booking will be cancelled (Sri Lankan hospitality context). The champion model is **XGBoost**; threshold based decisions currently use **0.35000000000000003** (see `champion_meta.json`).
|
| 32 |
|
| 33 |
-
_Last updated: 2025-10-05 16:
|
| 34 |
|
| 35 |
## Key Metrics (Holdout)
|
| 36 |
| Metric | Value |
|
|
@@ -42,7 +42,16 @@ _Last updated: 2025-10-05 16:16 UTC_
|
|
| 42 |
| Accuracy | 0.8613786749308987 |
|
| 43 |
|
| 44 |
## Top Features (SHAP importance)
|
| 45 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
|
| 47 |
## Quickstart
|
| 48 |
```python
|
|
|
|
| 30 |
|
| 31 |
Predicts probability that a hotel booking will be cancelled (Sri Lankan hospitality context). The champion model is **XGBoost**; threshold based decisions currently use **0.35000000000000003** (see `champion_meta.json`).
|
| 32 |
|
| 33 |
+
_Last updated: 2025-10-05 16:43 UTC_
|
| 34 |
|
| 35 |
## Key Metrics (Holdout)
|
| 36 |
| Metric | Value |
|
|
|
|
| 42 |
| Accuracy | 0.8613786749308987 |
|
| 43 |
|
| 44 |
## Top Features (SHAP importance)
|
| 45 |
+
- deposit_type
|
| 46 |
+
- country__te
|
| 47 |
+
- market_segment
|
| 48 |
+
- total_of_special_requests
|
| 49 |
+
- lead_time
|
| 50 |
+
- required_car_parking_spaces
|
| 51 |
+
- assigned_room_type
|
| 52 |
+
- customer_type_target_encoded
|
| 53 |
+
- reserved_room_type
|
| 54 |
+
- previous_cancellations
|
| 55 |
|
| 56 |
## Quickstart
|
| 57 |
```python
|