MinConfidence query strategy for multilabel classification. For more details on this query strategy, see Esuli and Sebastiani., Active Learning Strategies for Multi-Label Text Classification (http://dx.doi.org/10.1007/978-3-642-00958-7_12) Args: classifier: The multila
(classifier: OneVsRestClassifier, X_pool: modALinput,
n_instances: int = 1, random_tie_break: bool = False)
| 141 | |
| 142 | |
| 143 | def min_confidence(classifier: OneVsRestClassifier, X_pool: modALinput, |
| 144 | n_instances: int = 1, random_tie_break: bool = False) -> np.ndarray: |
| 145 | """ |
| 146 | MinConfidence query strategy for multilabel classification. |
| 147 | |
| 148 | For more details on this query strategy, see |
| 149 | Esuli and Sebastiani., Active Learning Strategies for Multi-Label Text Classification |
| 150 | (http://dx.doi.org/10.1007/978-3-642-00958-7_12) |
| 151 | |
| 152 | Args: |
| 153 | classifier: The multilabel classifier for which the labels are to be queried. |
| 154 | X_pool: The pool of samples to query from. |
| 155 | random_tie_break: If True, shuffles utility scores to randomize the order. This |
| 156 | can be used to break the tie when the highest utility score is not unique. |
| 157 | |
| 158 | Returns: |
| 159 | The index of the instance from X_pool chosen to be labelled. |
| 160 | The minimal confidence metric of the chosen instance. |
| 161 | |
| 162 | """ |
| 163 | |
| 164 | classwise_confidence = classifier.predict_proba(X_pool) |
| 165 | classwise_min = np.min(classwise_confidence, axis=1) |
| 166 | |
| 167 | if not random_tie_break: |
| 168 | return multi_argmin(classwise_min, n_instances) |
| 169 | |
| 170 | return shuffled_argmin(classwise_min, n_instances) |
| 171 | |
| 172 | |
| 173 | def avg_confidence(classifier: OneVsRestClassifier, X_pool: modALinput, |
nothing calls this directly
no test coverage detected
searching dependent graphs…