Use the trained model to classify or predict the examples in `X`. Parameters ---------- X : :py:class:`ndarray ` of shape `(N, M)` The training data of `N` examples, each with `M` features Returns ------- preds : :
(self, X)
| 156 | Y_pred[:, k] += self.weights[i, k] * h_pred |
| 157 | |
| 158 | def predict(self, X): |
| 159 | """ |
| 160 | Use the trained model to classify or predict the examples in `X`. |
| 161 | |
| 162 | Parameters |
| 163 | ---------- |
| 164 | X : :py:class:`ndarray <numpy.ndarray>` of shape `(N, M)` |
| 165 | The training data of `N` examples, each with `M` features |
| 166 | |
| 167 | Returns |
| 168 | ------- |
| 169 | preds : :py:class:`ndarray <numpy.ndarray>` of shape `(N,)` |
| 170 | The integer class labels predicted for each example in `X` if |
| 171 | ``self.classifier = True``, otherwise the predicted target values. |
| 172 | """ |
| 173 | Y_pred = np.zeros((X.shape[0], self.out_dims)) |
| 174 | for i in range(self.n_iter): |
| 175 | for k in range(self.out_dims): |
| 176 | Y_pred[:, k] += self.weights[i, k] * self.learners[i, k].predict(X) |
| 177 | |
| 178 | if self.classifier: |
| 179 | Y_pred = Y_pred.argmax(axis=1) |
| 180 | |
| 181 | return Y_pred |