Update README.md
Browse files
README.md
CHANGED
|
@@ -40,21 +40,9 @@ predictions, probabilities = clf.predict(X_train, y_train, X_test)
|
|
| 40 |
Labels can be any type (`int`, `str`, etc.).
|
| 41 |
Datasets with more than 10 classes are handled automatically via one-vs-rest.
|
| 42 |
|
| 43 |
-
## UCR Benchmark Results
|
| 44 |
-
|
| 45 |
-
| Dataset | Classes | Accuracy | ROC AUC |
|
| 46 |
-
|--------------------|---------|----------|---------|
|
| 47 |
-
| ArrowHead | 3 | 76.6 % | 0.964 |
|
| 48 |
-
| ECG5000 | 5 | 95.0 % | 0.952 |
|
| 49 |
-
| GunPoint | 2 | 98.7 % | 0.997 |
|
| 50 |
-
| ItalyPowerDemand | 2 | 96.0 % | 0.993 |
|
| 51 |
-
| TwoPatterns | 4 | 99.8 % | 1.000 |
|
| 52 |
-
|
| 53 |
-
Results use the default 4-member ensemble (interpolate × {256, 512} × {raw, first-difference}).
|
| 54 |
-
|
| 55 |
## Model Details
|
| 56 |
|
| 57 |
-
- **Parameters:** 4,
|
| 58 |
- **Input:** univariate or multivariate time series, any length
|
| 59 |
- **Output:** class probabilities over up to 10 classes (OvR for more)
|
| 60 |
- **Inference:** single forward pass; no dataset-specific adaptation
|
|
|
|
| 40 |
Labels can be any type (`int`, `str`, etc.).
|
| 41 |
Datasets with more than 10 classes are handled automatically via one-vs-rest.
|
| 42 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
## Model Details
|
| 44 |
|
| 45 |
+
- **Parameters:** ~4,5M
|
| 46 |
- **Input:** univariate or multivariate time series, any length
|
| 47 |
- **Output:** class probabilities over up to 10 classes (OvR for more)
|
| 48 |
- **Inference:** single forward pass; no dataset-specific adaptation
|