Return the predicted value for each sample. Parameters ---------- X : array-like or sparse matrix of shape = [n_samples, n_features] Input features matrix. raw_score : bool, optional (default=False) Whether to predict raw scores. num_i
(self, X, raw_score=False, num_iteration=None,
pred_leaf=False, pred_contrib=False, **kwargs)
| 620 | return self |
| 621 | |
| 622 | def predict(self, X, raw_score=False, num_iteration=None, |
| 623 | pred_leaf=False, pred_contrib=False, **kwargs): |
| 624 | """Return the predicted value for each sample. |
| 625 | |
| 626 | Parameters |
| 627 | ---------- |
| 628 | X : array-like or sparse matrix of shape = [n_samples, n_features] |
| 629 | Input features matrix. |
| 630 | raw_score : bool, optional (default=False) |
| 631 | Whether to predict raw scores. |
| 632 | num_iteration : int or None, optional (default=None) |
| 633 | Limit number of iterations in the prediction. |
| 634 | If None, if the best iteration exists, it is used; otherwise, all trees are used. |
| 635 | If <= 0, all trees are used (no limits). |
| 636 | pred_leaf : bool, optional (default=False) |
| 637 | Whether to predict leaf index. |
| 638 | pred_contrib : bool, optional (default=False) |
| 639 | Whether to predict feature contributions. |
| 640 | |
| 641 | .. note:: |
| 642 | |
| 643 | If you want to get more explanations for your model's predictions using SHAP values, |
| 644 | like SHAP interaction values, |
| 645 | you can install the shap package (https://github.com/slundberg/shap). |
| 646 | Note that unlike the shap package, with ``pred_contrib`` we return a matrix with an extra |
| 647 | column, where the last column is the expected value. |
| 648 | |
| 649 | **kwargs |
| 650 | Other parameters for the prediction. |
| 651 | |
| 652 | Returns |
| 653 | ------- |
| 654 | predicted_result : array-like of shape = [n_samples] or shape = [n_samples, n_classes] |
| 655 | The predicted values. |
| 656 | X_leaves : array-like of shape = [n_samples, n_trees] or shape = [n_samples, n_trees * n_classes] |
| 657 | If ``pred_leaf=True``, the predicted leaf of every tree for each sample. |
| 658 | X_SHAP_values : array-like of shape = [n_samples, n_features + 1] or shape = [n_samples, (n_features + 1) * n_classes] |
| 659 | If ``pred_contrib=True``, the feature contributions for each sample. |
| 660 | """ |
| 661 | if self._n_features is None: |
| 662 | raise LGBMNotFittedError("Estimator not fitted, call `fit` before exploiting the model.") |
| 663 | if not isinstance(X, (DataFrame, DataTable)): |
| 664 | X = _LGBMCheckArray(X, accept_sparse=True, force_all_finite=False) |
| 665 | n_features = X.shape[1] |
| 666 | if self._n_features != n_features: |
| 667 | raise ValueError("Number of features of the model must " |
| 668 | "match the input. Model n_features_ is %s and " |
| 669 | "input n_features is %s " |
| 670 | % (self._n_features, n_features)) |
| 671 | return self._Booster.predict(X, raw_score=raw_score, num_iteration=num_iteration, |
| 672 | pred_leaf=pred_leaf, pred_contrib=pred_contrib, **kwargs) |
| 673 | |
| 674 | @property |
| 675 | def n_features_(self): |