File size: 45,386 Bytes
7885a28 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 |
"""Stacking classifier and regressor."""
# Authors: The scikit-learn developers
# SPDX-License-Identifier: BSD-3-Clause
from abc import ABCMeta, abstractmethod
from copy import deepcopy
from numbers import Integral
import numpy as np
import scipy.sparse as sparse
from ..base import (
ClassifierMixin,
RegressorMixin,
TransformerMixin,
_fit_context,
clone,
is_classifier,
is_regressor,
)
from ..exceptions import NotFittedError
from ..linear_model import LogisticRegression, RidgeCV
from ..model_selection import check_cv, cross_val_predict
from ..preprocessing import LabelEncoder
from ..utils import Bunch
from ..utils._estimator_html_repr import _VisualBlock
from ..utils._param_validation import HasMethods, StrOptions
from ..utils.metadata_routing import (
MetadataRouter,
MethodMapping,
_raise_for_params,
_routing_enabled,
process_routing,
)
from ..utils.metaestimators import available_if
from ..utils.multiclass import check_classification_targets, type_of_target
from ..utils.parallel import Parallel, delayed
from ..utils.validation import (
_check_feature_names_in,
_check_response_method,
_deprecate_positional_args,
_estimator_has,
check_is_fitted,
column_or_1d,
)
from ._base import _BaseHeterogeneousEnsemble, _fit_single_estimator
class _BaseStacking(TransformerMixin, _BaseHeterogeneousEnsemble, metaclass=ABCMeta):
"""Base class for stacking method."""
_parameter_constraints: dict = {
"estimators": [list],
"final_estimator": [None, HasMethods("fit")],
"cv": ["cv_object", StrOptions({"prefit"})],
"n_jobs": [None, Integral],
"passthrough": ["boolean"],
"verbose": ["verbose"],
}
@abstractmethod
def __init__(
self,
estimators,
final_estimator=None,
*,
cv=None,
stack_method="auto",
n_jobs=None,
verbose=0,
passthrough=False,
):
super().__init__(estimators=estimators)
self.final_estimator = final_estimator
self.cv = cv
self.stack_method = stack_method
self.n_jobs = n_jobs
self.verbose = verbose
self.passthrough = passthrough
def _clone_final_estimator(self, default):
if self.final_estimator is not None:
self.final_estimator_ = clone(self.final_estimator)
else:
self.final_estimator_ = clone(default)
def _concatenate_predictions(self, X, predictions):
"""Concatenate the predictions of each first layer learner and
possibly the input dataset `X`.
If `X` is sparse and `self.passthrough` is False, the output of
`transform` will be dense (the predictions). If `X` is sparse
and `self.passthrough` is True, the output of `transform` will
be sparse.
This helper is in charge of ensuring the predictions are 2D arrays and
it will drop one of the probability column when using probabilities
in the binary case. Indeed, the p(y|c=0) = 1 - p(y|c=1)
When `y` type is `"multilabel-indicator"`` and the method used is
`predict_proba`, `preds` can be either a `ndarray` of shape
`(n_samples, n_class)` or for some estimators a list of `ndarray`.
This function will drop one of the probability column in this situation as well.
"""
X_meta = []
for est_idx, preds in enumerate(predictions):
if isinstance(preds, list):
# `preds` is here a list of `n_targets` 2D ndarrays of
# `n_classes` columns. The k-th column contains the
# probabilities of the samples belonging the k-th class.
#
# Since those probabilities must sum to one for each sample,
# we can work with probabilities of `n_classes - 1` classes.
# Hence we drop the first column.
for pred in preds:
X_meta.append(pred[:, 1:])
elif preds.ndim == 1:
# Some estimator return a 1D array for predictions
# which must be 2-dimensional arrays.
X_meta.append(preds.reshape(-1, 1))
elif (
self.stack_method_[est_idx] == "predict_proba"
and len(self.classes_) == 2
):
# Remove the first column when using probabilities in
# binary classification because both features `preds` are perfectly
# collinear.
X_meta.append(preds[:, 1:])
else:
X_meta.append(preds)
self._n_feature_outs = [pred.shape[1] for pred in X_meta]
if self.passthrough:
X_meta.append(X)
if sparse.issparse(X):
return sparse.hstack(X_meta, format=X.format)
return np.hstack(X_meta)
@staticmethod
def _method_name(name, estimator, method):
if estimator == "drop":
return None
if method == "auto":
method = ["predict_proba", "decision_function", "predict"]
try:
method_name = _check_response_method(estimator, method).__name__
except AttributeError as e:
raise ValueError(
f"Underlying estimator {name} does not implement the method {method}."
) from e
return method_name
@_fit_context(
# estimators in Stacking*.estimators are not validated yet
prefer_skip_nested_validation=False
)
def fit(self, X, y, **fit_params):
"""Fit the estimators.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
y : array-like of shape (n_samples,)
Target values.
**fit_params : dict
Dict of metadata, potentially containing sample_weight as a
key-value pair. If sample_weight is not present, then samples are
equally weighted. Note that sample_weight is supported only if all
underlying estimators support sample weights.
.. versionadded:: 1.6
Returns
-------
self : object
"""
# all_estimators contains all estimators, the one to be fitted and the
# 'drop' string.
names, all_estimators = self._validate_estimators()
self._validate_final_estimator()
stack_method = [self.stack_method] * len(all_estimators)
if _routing_enabled():
routed_params = process_routing(self, "fit", **fit_params)
else:
routed_params = Bunch()
for name in names:
routed_params[name] = Bunch(fit={})
if "sample_weight" in fit_params:
routed_params[name].fit["sample_weight"] = fit_params[
"sample_weight"
]
if self.cv == "prefit":
self.estimators_ = []
for estimator in all_estimators:
if estimator != "drop":
check_is_fitted(estimator)
self.estimators_.append(estimator)
else:
# Fit the base estimators on the whole training data. Those
# base estimators will be used in transform, predict, and
# predict_proba. They are exposed publicly.
self.estimators_ = Parallel(n_jobs=self.n_jobs)(
delayed(_fit_single_estimator)(
clone(est), X, y, routed_params[name]["fit"]
)
for name, est in zip(names, all_estimators)
if est != "drop"
)
self.named_estimators_ = Bunch()
est_fitted_idx = 0
for name_est, org_est in zip(names, all_estimators):
if org_est != "drop":
current_estimator = self.estimators_[est_fitted_idx]
self.named_estimators_[name_est] = current_estimator
est_fitted_idx += 1
if hasattr(current_estimator, "feature_names_in_"):
self.feature_names_in_ = current_estimator.feature_names_in_
else:
self.named_estimators_[name_est] = "drop"
self.stack_method_ = [
self._method_name(name, est, meth)
for name, est, meth in zip(names, all_estimators, stack_method)
]
if self.cv == "prefit":
# Generate predictions from prefit models
predictions = [
getattr(estimator, predict_method)(X)
for estimator, predict_method in zip(all_estimators, self.stack_method_)
if estimator != "drop"
]
else:
# To train the meta-classifier using the most data as possible, we use
# a cross-validation to obtain the output of the stacked estimators.
# To ensure that the data provided to each estimator are the same,
# we need to set the random state of the cv if there is one and we
# need to take a copy.
cv = check_cv(self.cv, y=y, classifier=is_classifier(self))
if hasattr(cv, "random_state") and cv.random_state is None:
cv.random_state = np.random.RandomState()
predictions = Parallel(n_jobs=self.n_jobs)(
delayed(cross_val_predict)(
clone(est),
X,
y,
cv=deepcopy(cv),
method=meth,
n_jobs=self.n_jobs,
params=routed_params[name]["fit"],
verbose=self.verbose,
)
for name, est, meth in zip(names, all_estimators, self.stack_method_)
if est != "drop"
)
# Only not None or not 'drop' estimators will be used in transform.
# Remove the None from the method as well.
self.stack_method_ = [
meth
for (meth, est) in zip(self.stack_method_, all_estimators)
if est != "drop"
]
X_meta = self._concatenate_predictions(X, predictions)
_fit_single_estimator(self.final_estimator_, X_meta, y, fit_params=fit_params)
return self
@property
def n_features_in_(self):
"""Number of features seen during :term:`fit`."""
try:
check_is_fitted(self)
except NotFittedError as nfe:
raise AttributeError(
f"{self.__class__.__name__} object has no attribute n_features_in_"
) from nfe
return self.estimators_[0].n_features_in_
def _transform(self, X):
"""Concatenate and return the predictions of the estimators."""
check_is_fitted(self)
predictions = [
getattr(est, meth)(X)
for est, meth in zip(self.estimators_, self.stack_method_)
if est != "drop"
]
return self._concatenate_predictions(X, predictions)
def get_feature_names_out(self, input_features=None):
"""Get output feature names for transformation.
Parameters
----------
input_features : array-like of str or None, default=None
Input features. The input feature names are only used when `passthrough` is
`True`.
- If `input_features` is `None`, then `feature_names_in_` is
used as feature names in. If `feature_names_in_` is not defined,
then names are generated: `[x0, x1, ..., x(n_features_in_ - 1)]`.
- If `input_features` is an array-like, then `input_features` must
match `feature_names_in_` if `feature_names_in_` is defined.
If `passthrough` is `False`, then only the names of `estimators` are used
to generate the output feature names.
Returns
-------
feature_names_out : ndarray of str objects
Transformed feature names.
"""
check_is_fitted(self, "n_features_in_")
input_features = _check_feature_names_in(
self, input_features, generate_names=self.passthrough
)
class_name = self.__class__.__name__.lower()
non_dropped_estimators = (
name for name, est in self.estimators if est != "drop"
)
meta_names = []
for est, n_features_out in zip(non_dropped_estimators, self._n_feature_outs):
if n_features_out == 1:
meta_names.append(f"{class_name}_{est}")
else:
meta_names.extend(
f"{class_name}_{est}{i}" for i in range(n_features_out)
)
if self.passthrough:
return np.concatenate((meta_names, input_features))
return np.asarray(meta_names, dtype=object)
@available_if(
_estimator_has("predict", delegates=("final_estimator_", "final_estimator"))
)
def predict(self, X, **predict_params):
"""Predict target for X.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
**predict_params : dict of str -> obj
Parameters to the `predict` called by the `final_estimator`. Note
that this may be used to return uncertainties from some estimators
with `return_std` or `return_cov`. Be aware that it will only
account for uncertainty in the final estimator.
Returns
-------
y_pred : ndarray of shape (n_samples,) or (n_samples, n_output)
Predicted targets.
"""
check_is_fitted(self)
return self.final_estimator_.predict(self.transform(X), **predict_params)
def _sk_visual_block_with_final_estimator(self, final_estimator):
names, estimators = zip(*self.estimators)
parallel = _VisualBlock("parallel", estimators, names=names, dash_wrapped=False)
# final estimator is wrapped in a parallel block to show the label:
# 'final_estimator' in the html repr
final_block = _VisualBlock(
"parallel", [final_estimator], names=["final_estimator"], dash_wrapped=False
)
return _VisualBlock("serial", (parallel, final_block), dash_wrapped=False)
def get_metadata_routing(self):
"""Get metadata routing of this object.
Please check :ref:`User Guide <metadata_routing>` on how the routing
mechanism works.
.. versionadded:: 1.6
Returns
-------
routing : MetadataRouter
A :class:`~sklearn.utils.metadata_routing.MetadataRouter` encapsulating
routing information.
"""
router = MetadataRouter(owner=self.__class__.__name__)
# `self.estimators` is a list of (name, est) tuples
for name, estimator in self.estimators:
router.add(
**{name: estimator},
method_mapping=MethodMapping().add(callee="fit", caller="fit"),
)
try:
final_estimator_ = self.final_estimator_
except AttributeError:
final_estimator_ = self.final_estimator
router.add(
final_estimator_=final_estimator_,
method_mapping=MethodMapping().add(caller="predict", callee="predict"),
)
return router
class StackingClassifier(ClassifierMixin, _BaseStacking):
"""Stack of estimators with a final classifier.
Stacked generalization consists in stacking the output of individual
estimator and use a classifier to compute the final prediction. Stacking
allows to use the strength of each individual estimator by using their
output as input of a final estimator.
Note that `estimators_` are fitted on the full `X` while `final_estimator_`
is trained using cross-validated predictions of the base estimators using
`cross_val_predict`.
Read more in the :ref:`User Guide <stacking>`.
.. versionadded:: 0.22
Parameters
----------
estimators : list of (str, estimator)
Base estimators which will be stacked together. Each element of the
list is defined as a tuple of string (i.e. name) and an estimator
instance. An estimator can be set to 'drop' using `set_params`.
The type of estimator is generally expected to be a classifier.
However, one can pass a regressor for some use case (e.g. ordinal
regression).
final_estimator : estimator, default=None
A classifier which will be used to combine the base estimators.
The default classifier is a
:class:`~sklearn.linear_model.LogisticRegression`.
cv : int, cross-validation generator, iterable, or "prefit", default=None
Determines the cross-validation splitting strategy used in
`cross_val_predict` to train `final_estimator`. Possible inputs for
cv are:
* None, to use the default 5-fold cross validation,
* integer, to specify the number of folds in a (Stratified) KFold,
* An object to be used as a cross-validation generator,
* An iterable yielding train, test splits,
* `"prefit"`, to assume the `estimators` are prefit. In this case, the
estimators will not be refitted.
For integer/None inputs, if the estimator is a classifier and y is
either binary or multiclass,
:class:`~sklearn.model_selection.StratifiedKFold` is used.
In all other cases, :class:`~sklearn.model_selection.KFold` is used.
These splitters are instantiated with `shuffle=False` so the splits
will be the same across calls.
Refer :ref:`User Guide <cross_validation>` for the various
cross-validation strategies that can be used here.
If "prefit" is passed, it is assumed that all `estimators` have
been fitted already. The `final_estimator_` is trained on the `estimators`
predictions on the full training set and are **not** cross validated
predictions. Please note that if the models have been trained on the same
data to train the stacking model, there is a very high risk of overfitting.
.. versionadded:: 1.1
The 'prefit' option was added in 1.1
.. note::
A larger number of split will provide no benefits if the number
of training samples is large enough. Indeed, the training time
will increase. ``cv`` is not used for model evaluation but for
prediction.
stack_method : {'auto', 'predict_proba', 'decision_function', 'predict'}, \
default='auto'
Methods called for each base estimator. It can be:
* if 'auto', it will try to invoke, for each estimator,
`'predict_proba'`, `'decision_function'` or `'predict'` in that
order.
* otherwise, one of `'predict_proba'`, `'decision_function'` or
`'predict'`. If the method is not implemented by the estimator, it
will raise an error.
n_jobs : int, default=None
The number of jobs to run in parallel for `fit` of all `estimators`.
`None` means 1 unless in a `joblib.parallel_backend` context. -1 means
using all processors. See :term:`Glossary <n_jobs>` for more details.
passthrough : bool, default=False
When False, only the predictions of estimators will be used as
training data for `final_estimator`. When True, the
`final_estimator` is trained on the predictions as well as the
original training data.
verbose : int, default=0
Verbosity level.
Attributes
----------
classes_ : ndarray of shape (n_classes,) or list of ndarray if `y` \
is of type `"multilabel-indicator"`.
Class labels.
estimators_ : list of estimators
The elements of the `estimators` parameter, having been fitted on the
training data. If an estimator has been set to `'drop'`, it
will not appear in `estimators_`. When `cv="prefit"`, `estimators_`
is set to `estimators` and is not fitted again.
named_estimators_ : :class:`~sklearn.utils.Bunch`
Attribute to access any fitted sub-estimators by name.
n_features_in_ : int
Number of features seen during :term:`fit`. Only defined if the
underlying estimator exposes such an attribute when fit.
.. versionadded:: 0.24
feature_names_in_ : ndarray of shape (`n_features_in_`,)
Names of features seen during :term:`fit`. Only defined if the
underlying estimators expose such an attribute when fit.
.. versionadded:: 1.0
final_estimator_ : estimator
The classifier fit on the output of `estimators_` and responsible for
final predictions.
stack_method_ : list of str
The method used by each base estimator.
See Also
--------
StackingRegressor : Stack of estimators with a final regressor.
Notes
-----
When `predict_proba` is used by each estimator (i.e. most of the time for
`stack_method='auto'` or specifically for `stack_method='predict_proba'`),
the first column predicted by each estimator will be dropped in the case
of a binary classification problem. Indeed, both feature will be perfectly
collinear.
In some cases (e.g. ordinal regression), one can pass regressors as the
first layer of the :class:`StackingClassifier`. However, note that `y` will
be internally encoded in a numerically increasing order or lexicographic
order. If this ordering is not adequate, one should manually numerically
encode the classes in the desired order.
References
----------
.. [1] Wolpert, David H. "Stacked generalization." Neural networks 5.2
(1992): 241-259.
Examples
--------
>>> from sklearn.datasets import load_iris
>>> from sklearn.ensemble import RandomForestClassifier
>>> from sklearn.svm import LinearSVC
>>> from sklearn.linear_model import LogisticRegression
>>> from sklearn.preprocessing import StandardScaler
>>> from sklearn.pipeline import make_pipeline
>>> from sklearn.ensemble import StackingClassifier
>>> X, y = load_iris(return_X_y=True)
>>> estimators = [
... ('rf', RandomForestClassifier(n_estimators=10, random_state=42)),
... ('svr', make_pipeline(StandardScaler(),
... LinearSVC(random_state=42)))
... ]
>>> clf = StackingClassifier(
... estimators=estimators, final_estimator=LogisticRegression()
... )
>>> from sklearn.model_selection import train_test_split
>>> X_train, X_test, y_train, y_test = train_test_split(
... X, y, stratify=y, random_state=42
... )
>>> clf.fit(X_train, y_train).score(X_test, y_test)
0.9...
"""
_parameter_constraints: dict = {
**_BaseStacking._parameter_constraints,
"stack_method": [
StrOptions({"auto", "predict_proba", "decision_function", "predict"})
],
}
def __init__(
self,
estimators,
final_estimator=None,
*,
cv=None,
stack_method="auto",
n_jobs=None,
passthrough=False,
verbose=0,
):
super().__init__(
estimators=estimators,
final_estimator=final_estimator,
cv=cv,
stack_method=stack_method,
n_jobs=n_jobs,
passthrough=passthrough,
verbose=verbose,
)
def _validate_final_estimator(self):
self._clone_final_estimator(default=LogisticRegression())
if not is_classifier(self.final_estimator_):
raise ValueError(
"'final_estimator' parameter should be a classifier. Got {}".format(
self.final_estimator_
)
)
def _validate_estimators(self):
"""Overload the method of `_BaseHeterogeneousEnsemble` to be more
lenient towards the type of `estimators`.
Regressors can be accepted for some cases such as ordinal regression.
"""
if len(self.estimators) == 0:
raise ValueError(
"Invalid 'estimators' attribute, 'estimators' should be a "
"non-empty list of (string, estimator) tuples."
)
names, estimators = zip(*self.estimators)
self._validate_names(names)
has_estimator = any(est != "drop" for est in estimators)
if not has_estimator:
raise ValueError(
"All estimators are dropped. At least one is required "
"to be an estimator."
)
return names, estimators
# TODO(1.7): remove `sample_weight` from the signature after deprecation
# cycle; pop it from `fit_params` before the `_raise_for_params` check and
# reinsert afterwards, for backwards compatibility
@_deprecate_positional_args(version="1.7")
def fit(self, X, y, *, sample_weight=None, **fit_params):
"""Fit the estimators.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
y : array-like of shape (n_samples,)
Target values. Note that `y` will be internally encoded in
numerically increasing order or lexicographic order. If the order
matter (e.g. for ordinal regression), one should numerically encode
the target `y` before calling :term:`fit`.
sample_weight : array-like of shape (n_samples,), default=None
Sample weights. If None, then samples are equally weighted.
Note that this is supported only if all underlying estimators
support sample weights.
**fit_params : dict
Parameters to pass to the underlying estimators.
.. versionadded:: 1.6
Only available if `enable_metadata_routing=True`, which can be
set by using ``sklearn.set_config(enable_metadata_routing=True)``.
See :ref:`Metadata Routing User Guide <metadata_routing>` for
more details.
Returns
-------
self : object
Returns a fitted instance of estimator.
"""
_raise_for_params(fit_params, self, "fit")
check_classification_targets(y)
if type_of_target(y) == "multilabel-indicator":
self._label_encoder = [LabelEncoder().fit(yk) for yk in y.T]
self.classes_ = [le.classes_ for le in self._label_encoder]
y_encoded = np.array(
[
self._label_encoder[target_idx].transform(target)
for target_idx, target in enumerate(y.T)
]
).T
else:
self._label_encoder = LabelEncoder().fit(y)
self.classes_ = self._label_encoder.classes_
y_encoded = self._label_encoder.transform(y)
if sample_weight is not None:
fit_params["sample_weight"] = sample_weight
return super().fit(X, y_encoded, **fit_params)
@available_if(
_estimator_has("predict", delegates=("final_estimator_", "final_estimator"))
)
def predict(self, X, **predict_params):
"""Predict target for X.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
**predict_params : dict of str -> obj
Parameters to the `predict` called by the `final_estimator`. Note
that this may be used to return uncertainties from some estimators
with `return_std` or `return_cov`. Be aware that it will only
account for uncertainty in the final estimator.
- If `enable_metadata_routing=False` (default):
Parameters directly passed to the `predict` method of the
`final_estimator`.
- If `enable_metadata_routing=True`: Parameters safely routed to
the `predict` method of the `final_estimator`. See :ref:`Metadata
Routing User Guide <metadata_routing>` for more details.
.. versionchanged:: 1.6
`**predict_params` can be routed via metadata routing API.
Returns
-------
y_pred : ndarray of shape (n_samples,) or (n_samples, n_output)
Predicted targets.
"""
if _routing_enabled():
routed_params = process_routing(self, "predict", **predict_params)
else:
# TODO(SLEP6): remove when metadata routing cannot be disabled.
routed_params = Bunch()
routed_params.final_estimator_ = Bunch(predict={})
routed_params.final_estimator_.predict = predict_params
y_pred = super().predict(X, **routed_params.final_estimator_["predict"])
if isinstance(self._label_encoder, list):
# Handle the multilabel-indicator case
y_pred = np.array(
[
self._label_encoder[target_idx].inverse_transform(target)
for target_idx, target in enumerate(y_pred.T)
]
).T
else:
y_pred = self._label_encoder.inverse_transform(y_pred)
return y_pred
@available_if(
_estimator_has(
"predict_proba", delegates=("final_estimator_", "final_estimator")
)
)
def predict_proba(self, X):
"""Predict class probabilities for `X` using the final estimator.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
Returns
-------
probabilities : ndarray of shape (n_samples, n_classes) or \
list of ndarray of shape (n_output,)
The class probabilities of the input samples.
"""
check_is_fitted(self)
y_pred = self.final_estimator_.predict_proba(self.transform(X))
if isinstance(self._label_encoder, list):
# Handle the multilabel-indicator cases
y_pred = np.array([preds[:, 0] for preds in y_pred]).T
return y_pred
@available_if(
_estimator_has(
"decision_function", delegates=("final_estimator_", "final_estimator")
)
)
def decision_function(self, X):
"""Decision function for samples in `X` using the final estimator.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
Returns
-------
decisions : ndarray of shape (n_samples,), (n_samples, n_classes), \
or (n_samples, n_classes * (n_classes-1) / 2)
The decision function computed the final estimator.
"""
check_is_fitted(self)
return self.final_estimator_.decision_function(self.transform(X))
def transform(self, X):
"""Return class labels or probabilities for X for each estimator.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
Returns
-------
y_preds : ndarray of shape (n_samples, n_estimators) or \
(n_samples, n_classes * n_estimators)
Prediction outputs for each estimator.
"""
return self._transform(X)
def _sk_visual_block_(self):
# If final_estimator's default changes then this should be
# updated.
if self.final_estimator is None:
final_estimator = LogisticRegression()
else:
final_estimator = self.final_estimator
return super()._sk_visual_block_with_final_estimator(final_estimator)
class StackingRegressor(RegressorMixin, _BaseStacking):
"""Stack of estimators with a final regressor.
Stacked generalization consists in stacking the output of individual
estimator and use a regressor to compute the final prediction. Stacking
allows to use the strength of each individual estimator by using their
output as input of a final estimator.
Note that `estimators_` are fitted on the full `X` while `final_estimator_`
is trained using cross-validated predictions of the base estimators using
`cross_val_predict`.
Read more in the :ref:`User Guide <stacking>`.
.. versionadded:: 0.22
Parameters
----------
estimators : list of (str, estimator)
Base estimators which will be stacked together. Each element of the
list is defined as a tuple of string (i.e. name) and an estimator
instance. An estimator can be set to 'drop' using `set_params`.
final_estimator : estimator, default=None
A regressor which will be used to combine the base estimators.
The default regressor is a :class:`~sklearn.linear_model.RidgeCV`.
cv : int, cross-validation generator, iterable, or "prefit", default=None
Determines the cross-validation splitting strategy used in
`cross_val_predict` to train `final_estimator`. Possible inputs for
cv are:
* None, to use the default 5-fold cross validation,
* integer, to specify the number of folds in a (Stratified) KFold,
* An object to be used as a cross-validation generator,
* An iterable yielding train, test splits,
* `"prefit"`, to assume the `estimators` are prefit. In this case, the
estimators will not be refitted.
For integer/None inputs, if the estimator is a classifier and y is
either binary or multiclass,
:class:`~sklearn.model_selection.StratifiedKFold` is used.
In all other cases, :class:`~sklearn.model_selection.KFold` is used.
These splitters are instantiated with `shuffle=False` so the splits
will be the same across calls.
Refer :ref:`User Guide <cross_validation>` for the various
cross-validation strategies that can be used here.
If "prefit" is passed, it is assumed that all `estimators` have
been fitted already. The `final_estimator_` is trained on the `estimators`
predictions on the full training set and are **not** cross validated
predictions. Please note that if the models have been trained on the same
data to train the stacking model, there is a very high risk of overfitting.
.. versionadded:: 1.1
The 'prefit' option was added in 1.1
.. note::
A larger number of split will provide no benefits if the number
of training samples is large enough. Indeed, the training time
will increase. ``cv`` is not used for model evaluation but for
prediction.
n_jobs : int, default=None
The number of jobs to run in parallel for `fit` of all `estimators`.
`None` means 1 unless in a `joblib.parallel_backend` context. -1 means
using all processors. See :term:`Glossary <n_jobs>` for more details.
passthrough : bool, default=False
When False, only the predictions of estimators will be used as
training data for `final_estimator`. When True, the
`final_estimator` is trained on the predictions as well as the
original training data.
verbose : int, default=0
Verbosity level.
Attributes
----------
estimators_ : list of estimators
The elements of the `estimators` parameter, having been fitted on the
training data. If an estimator has been set to `'drop'`, it
will not appear in `estimators_`. When `cv="prefit"`, `estimators_`
is set to `estimators` and is not fitted again.
named_estimators_ : :class:`~sklearn.utils.Bunch`
Attribute to access any fitted sub-estimators by name.
n_features_in_ : int
Number of features seen during :term:`fit`. Only defined if the
underlying estimator exposes such an attribute when fit.
.. versionadded:: 0.24
feature_names_in_ : ndarray of shape (`n_features_in_`,)
Names of features seen during :term:`fit`. Only defined if the
underlying estimators expose such an attribute when fit.
.. versionadded:: 1.0
final_estimator_ : estimator
The regressor fit on the output of `estimators_` and responsible for
final predictions.
stack_method_ : list of str
The method used by each base estimator.
See Also
--------
StackingClassifier : Stack of estimators with a final classifier.
References
----------
.. [1] Wolpert, David H. "Stacked generalization." Neural networks 5.2
(1992): 241-259.
Examples
--------
>>> from sklearn.datasets import load_diabetes
>>> from sklearn.linear_model import RidgeCV
>>> from sklearn.svm import LinearSVR
>>> from sklearn.ensemble import RandomForestRegressor
>>> from sklearn.ensemble import StackingRegressor
>>> X, y = load_diabetes(return_X_y=True)
>>> estimators = [
... ('lr', RidgeCV()),
... ('svr', LinearSVR(random_state=42))
... ]
>>> reg = StackingRegressor(
... estimators=estimators,
... final_estimator=RandomForestRegressor(n_estimators=10,
... random_state=42)
... )
>>> from sklearn.model_selection import train_test_split
>>> X_train, X_test, y_train, y_test = train_test_split(
... X, y, random_state=42
... )
>>> reg.fit(X_train, y_train).score(X_test, y_test)
0.3...
"""
def __init__(
self,
estimators,
final_estimator=None,
*,
cv=None,
n_jobs=None,
passthrough=False,
verbose=0,
):
super().__init__(
estimators=estimators,
final_estimator=final_estimator,
cv=cv,
stack_method="predict",
n_jobs=n_jobs,
passthrough=passthrough,
verbose=verbose,
)
def _validate_final_estimator(self):
self._clone_final_estimator(default=RidgeCV())
if not is_regressor(self.final_estimator_):
raise ValueError(
"'final_estimator' parameter should be a regressor. Got {}".format(
self.final_estimator_
)
)
# TODO(1.7): remove `sample_weight` from the signature after deprecation
# cycle; pop it from `fit_params` before the `_raise_for_params` check and
# reinsert afterwards, for backwards compatibility
@_deprecate_positional_args(version="1.7")
def fit(self, X, y, *, sample_weight=None, **fit_params):
"""Fit the estimators.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
y : array-like of shape (n_samples,)
Target values.
sample_weight : array-like of shape (n_samples,), default=None
Sample weights. If None, then samples are equally weighted.
Note that this is supported only if all underlying estimators
support sample weights.
**fit_params : dict
Parameters to pass to the underlying estimators.
.. versionadded:: 1.6
Only available if `enable_metadata_routing=True`, which can be
set by using ``sklearn.set_config(enable_metadata_routing=True)``.
See :ref:`Metadata Routing User Guide <metadata_routing>` for
more details.
Returns
-------
self : object
Returns a fitted instance.
"""
_raise_for_params(fit_params, self, "fit")
y = column_or_1d(y, warn=True)
if sample_weight is not None:
fit_params["sample_weight"] = sample_weight
return super().fit(X, y, **fit_params)
def transform(self, X):
"""Return the predictions for X for each estimator.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
Returns
-------
y_preds : ndarray of shape (n_samples, n_estimators)
Prediction outputs for each estimator.
"""
return self._transform(X)
# TODO(1.7): remove `sample_weight` from the signature after deprecation
# cycle; pop it from `fit_params` before the `_raise_for_params` check and
# reinsert afterwards, for backwards compatibility
@_deprecate_positional_args(version="1.7")
def fit_transform(self, X, y, *, sample_weight=None, **fit_params):
"""Fit the estimators and return the predictions for X for each estimator.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
y : array-like of shape (n_samples,)
Target values.
sample_weight : array-like of shape (n_samples,), default=None
Sample weights. If None, then samples are equally weighted.
Note that this is supported only if all underlying estimators
support sample weights.
**fit_params : dict
Parameters to pass to the underlying estimators.
.. versionadded:: 1.6
Only available if `enable_metadata_routing=True`, which can be
set by using ``sklearn.set_config(enable_metadata_routing=True)``.
See :ref:`Metadata Routing User Guide <metadata_routing>` for
more details.
Returns
-------
y_preds : ndarray of shape (n_samples, n_estimators)
Prediction outputs for each estimator.
"""
_raise_for_params(fit_params, self, "fit")
if sample_weight is not None:
fit_params["sample_weight"] = sample_weight
return super().fit_transform(X, y, **fit_params)
@available_if(
_estimator_has("predict", delegates=("final_estimator_", "final_estimator"))
)
def predict(self, X, **predict_params):
"""Predict target for X.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Training vectors, where `n_samples` is the number of samples and
`n_features` is the number of features.
**predict_params : dict of str -> obj
Parameters to the `predict` called by the `final_estimator`. Note
that this may be used to return uncertainties from some estimators
with `return_std` or `return_cov`. Be aware that it will only
account for uncertainty in the final estimator.
- If `enable_metadata_routing=False` (default):
Parameters directly passed to the `predict` method of the
`final_estimator`.
- If `enable_metadata_routing=True`: Parameters safely routed to
the `predict` method of the `final_estimator`. See :ref:`Metadata
Routing User Guide <metadata_routing>` for more details.
.. versionchanged:: 1.6
`**predict_params` can be routed via metadata routing API.
Returns
-------
y_pred : ndarray of shape (n_samples,) or (n_samples, n_output)
Predicted targets.
"""
if _routing_enabled():
routed_params = process_routing(self, "predict", **predict_params)
else:
# TODO(SLEP6): remove when metadata routing cannot be disabled.
routed_params = Bunch()
routed_params.final_estimator_ = Bunch(predict={})
routed_params.final_estimator_.predict = predict_params
y_pred = super().predict(X, **routed_params.final_estimator_["predict"])
return y_pred
def _sk_visual_block_(self):
# If final_estimator's default changes then this should be
# updated.
if self.final_estimator is None:
final_estimator = RidgeCV()
else:
final_estimator = self.final_estimator
return super()._sk_visual_block_with_final_estimator(final_estimator)
|