Trains the predictor with the passed data (warm_start decides if params are resetted or not). Args: X: The new samples for which the labels are supplied by the expert. y: Labels corresponding to the new instances in X. warm_start: If False, the
(self, X: modALinput, y: modALinput, warm_start: bool = True, bootstrap: bool = False, **fit_kwargs)
| 229 | return self._fit_on_new(X, y, bootstrap=bootstrap, **fit_kwargs) |
| 230 | |
| 231 | def teach(self, X: modALinput, y: modALinput, warm_start: bool = True, bootstrap: bool = False, **fit_kwargs) -> None: |
| 232 | """ |
| 233 | Trains the predictor with the passed data (warm_start decides if params are resetted or not). |
| 234 | |
| 235 | Args: |
| 236 | X: The new samples for which the labels are supplied by the expert. |
| 237 | y: Labels corresponding to the new instances in X. |
| 238 | warm_start: If False, the model parameters are resetted and the training starts from zero, |
| 239 | otherwise the pre trained model is kept and further trained. |
| 240 | bootstrap: If True, training is done on a bootstrapped dataset. Useful for building Committee models |
| 241 | with bagging. |
| 242 | **fit_kwargs: Keyword arguments to be passed to the fit method of the predictor. |
| 243 | """ |
| 244 | |
| 245 | if warm_start: |
| 246 | if not bootstrap: |
| 247 | self.estimator.partial_fit(X, y, **fit_kwargs) |
| 248 | else: |
| 249 | bootstrap_idx = np.random.choice( |
| 250 | range(X.shape[0]), X.shape[0], replace=True) |
| 251 | self.estimator.partial_fit( |
| 252 | X[bootstrap_idx], y[bootstrap_idx], **fit_kwargs) |
| 253 | else: |
| 254 | self._fit_on_new(X, y, bootstrap=bootstrap, **fit_kwargs) |
| 255 | |
| 256 | @property |
| 257 | def num_epochs(self): |