-
-
Notifications
You must be signed in to change notification settings - Fork 25.3k
/
_univariate_selection.py
1161 lines (924 loc) · 39.4 KB
/
_univariate_selection.py
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
"""Univariate features selection."""
# Authors: V. Michel, B. Thirion, G. Varoquaux, A. Gramfort, E. Duchesnay.
# L. Buitinck, A. Joly
# License: BSD 3 clause
import warnings
from numbers import Integral, Real
import numpy as np
from scipy import special, stats
from scipy.sparse import issparse
from ..base import BaseEstimator, _fit_context
from ..preprocessing import LabelBinarizer
from ..utils import as_float_array, check_array, check_X_y, safe_mask, safe_sqr
from ..utils._param_validation import Interval, StrOptions, validate_params
from ..utils.extmath import row_norms, safe_sparse_dot
from ..utils.validation import check_is_fitted
from ._base import SelectorMixin
def _clean_nans(scores):
"""
Fixes Issue #1240: NaNs can't be properly compared, so change them to the
smallest value of scores's dtype. -inf seems to be unreliable.
"""
# XXX where should this function be called? fit? scoring functions
# themselves?
scores = as_float_array(scores, copy=True)
scores[np.isnan(scores)] = np.finfo(scores.dtype).min
return scores
######################################################################
# Scoring functions
# The following function is a rewriting of scipy.stats.f_oneway
# Contrary to the scipy.stats.f_oneway implementation it does not
# copy the data while keeping the inputs unchanged.
def f_oneway(*args):
"""Perform a 1-way ANOVA.
The one-way ANOVA tests the null hypothesis that 2 or more groups have
the same population mean. The test is applied to samples from two or
more groups, possibly with differing sizes.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
*args : {array-like, sparse matrix}
Sample1, sample2... The sample measurements should be given as
arguments.
Returns
-------
f_statistic : float
The computed F-value of the test.
p_value : float
The associated p-value from the F-distribution.
Notes
-----
The ANOVA test has important assumptions that must be satisfied in order
for the associated p-value to be valid.
1. The samples are independent
2. Each sample is from a normally distributed population
3. The population standard deviations of the groups are all equal. This
property is known as homoscedasticity.
If these assumptions are not true for a given set of data, it may still be
possible to use the Kruskal-Wallis H-test (`scipy.stats.kruskal`_) although
with some loss of power.
The algorithm is from Heiman[2], pp.394-7.
See ``scipy.stats.f_oneway`` that should give the same results while
being less efficient.
References
----------
.. [1] Lowry, Richard. "Concepts and Applications of Inferential
Statistics". Chapter 14.
http://vassarstats.net/textbook
.. [2] Heiman, G.W. Research Methods in Statistics. 2002.
"""
n_classes = len(args)
args = [as_float_array(a) for a in args]
n_samples_per_class = np.array([a.shape[0] for a in args])
n_samples = np.sum(n_samples_per_class)
ss_alldata = sum(safe_sqr(a).sum(axis=0) for a in args)
sums_args = [np.asarray(a.sum(axis=0)) for a in args]
square_of_sums_alldata = sum(sums_args) ** 2
square_of_sums_args = [s**2 for s in sums_args]
sstot = ss_alldata - square_of_sums_alldata / float(n_samples)
ssbn = 0.0
for k, _ in enumerate(args):
ssbn += square_of_sums_args[k] / n_samples_per_class[k]
ssbn -= square_of_sums_alldata / float(n_samples)
sswn = sstot - ssbn
dfbn = n_classes - 1
dfwn = n_samples - n_classes
msb = ssbn / float(dfbn)
msw = sswn / float(dfwn)
constant_features_idx = np.where(msw == 0.0)[0]
if np.nonzero(msb)[0].size != msb.size and constant_features_idx.size:
warnings.warn("Features %s are constant." % constant_features_idx, UserWarning)
f = msb / msw
# flatten matrix to vector in sparse case
f = np.asarray(f).ravel()
prob = special.fdtrc(dfbn, dfwn, f)
return f, prob
@validate_params(
{
"X": ["array-like", "sparse matrix"],
"y": ["array-like"],
},
prefer_skip_nested_validation=True,
)
def f_classif(X, y):
"""Compute the ANOVA F-value for the provided sample.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
The set of regressors that will be tested sequentially.
y : array-like of shape (n_samples,)
The target vector.
Returns
-------
f_statistic : ndarray of shape (n_features,)
F-statistic for each feature.
p_values : ndarray of shape (n_features,)
P-values associated with the F-statistic.
See Also
--------
chi2 : Chi-squared stats of non-negative features for classification tasks.
f_regression : F-value between label/feature for regression tasks.
Examples
--------
>>> from sklearn.datasets import make_classification
>>> from sklearn.feature_selection import f_classif
>>> X, y = make_classification(
... n_samples=100, n_features=10, n_informative=2, n_clusters_per_class=1,
... shuffle=False, random_state=42
... )
>>> f_statistic, p_values = f_classif(X, y)
>>> f_statistic
array([2.2...e+02, 7.0...e-01, 1.6...e+00, 9.3...e-01,
5.4...e+00, 3.2...e-01, 4.7...e-02, 5.7...e-01,
7.5...e-01, 8.9...e-02])
>>> p_values
array([7.1...e-27, 4.0...e-01, 1.9...e-01, 3.3...e-01,
2.2...e-02, 5.7...e-01, 8.2...e-01, 4.5...e-01,
3.8...e-01, 7.6...e-01])
"""
X, y = check_X_y(X, y, accept_sparse=["csr", "csc", "coo"])
args = [X[safe_mask(X, y == k)] for k in np.unique(y)]
return f_oneway(*args)
def _chisquare(f_obs, f_exp):
"""Fast replacement for scipy.stats.chisquare.
Version from https://github.com/scipy/scipy/pull/2525 with additional
optimizations.
"""
f_obs = np.asarray(f_obs, dtype=np.float64)
k = len(f_obs)
# Reuse f_obs for chi-squared statistics
chisq = f_obs
chisq -= f_exp
chisq **= 2
with np.errstate(invalid="ignore"):
chisq /= f_exp
chisq = chisq.sum(axis=0)
return chisq, special.chdtrc(k - 1, chisq)
@validate_params(
{
"X": ["array-like", "sparse matrix"],
"y": ["array-like"],
},
prefer_skip_nested_validation=True,
)
def chi2(X, y):
"""Compute chi-squared stats between each non-negative feature and class.
This score can be used to select the `n_features` features with the
highest values for the test chi-squared statistic from X, which must
contain only **non-negative features** such as booleans or frequencies
(e.g., term counts in document classification), relative to the classes.
Recall that the chi-square test measures dependence between stochastic
variables, so using this function "weeds out" the features that are the
most likely to be independent of class and therefore irrelevant for
classification.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
Sample vectors.
y : array-like of shape (n_samples,)
Target vector (class labels).
Returns
-------
chi2 : ndarray of shape (n_features,)
Chi2 statistics for each feature.
p_values : ndarray of shape (n_features,)
P-values for each feature.
See Also
--------
f_classif : ANOVA F-value between label/feature for classification tasks.
f_regression : F-value between label/feature for regression tasks.
Notes
-----
Complexity of this algorithm is O(n_classes * n_features).
Examples
--------
>>> import numpy as np
>>> from sklearn.feature_selection import chi2
>>> X = np.array([[1, 1, 3],
... [0, 1, 5],
... [5, 4, 1],
... [6, 6, 2],
... [1, 4, 0],
... [0, 0, 0]])
>>> y = np.array([1, 1, 0, 0, 2, 2])
>>> chi2_stats, p_values = chi2(X, y)
>>> chi2_stats
array([15.3..., 6.5 , 8.9...])
>>> p_values
array([0.0004..., 0.0387..., 0.0116... ])
"""
# XXX: we might want to do some of the following in logspace instead for
# numerical stability.
# Converting X to float allows getting better performance for the
# safe_sparse_dot call made below.
X = check_array(X, accept_sparse="csr", dtype=(np.float64, np.float32))
if np.any((X.data if issparse(X) else X) < 0):
raise ValueError("Input X must be non-negative.")
# Use a sparse representation for Y by default to reduce memory usage when
# y has many unique classes.
Y = LabelBinarizer(sparse_output=True).fit_transform(y)
if Y.shape[1] == 1:
Y = Y.toarray()
Y = np.append(1 - Y, Y, axis=1)
observed = safe_sparse_dot(Y.T, X) # n_classes * n_features
if issparse(observed):
# convert back to a dense array before calling _chisquare
# XXX: could _chisquare be reimplement to accept sparse matrices for
# cases where both n_classes and n_features are large (and X is
# sparse)?
observed = observed.toarray()
feature_count = X.sum(axis=0).reshape(1, -1)
class_prob = Y.mean(axis=0).reshape(1, -1)
expected = np.dot(class_prob.T, feature_count)
return _chisquare(observed, expected)
@validate_params(
{
"X": ["array-like", "sparse matrix"],
"y": ["array-like"],
"center": ["boolean"],
"force_finite": ["boolean"],
},
prefer_skip_nested_validation=True,
)
def r_regression(X, y, *, center=True, force_finite=True):
"""Compute Pearson's r for each features and the target.
Pearson's r is also known as the Pearson correlation coefficient.
Linear model for testing the individual effect of each of many regressors.
This is a scoring function to be used in a feature selection procedure, not
a free standing feature selection procedure.
The cross correlation between each regressor and the target is computed
as::
E[(X[:, i] - mean(X[:, i])) * (y - mean(y))] / (std(X[:, i]) * std(y))
For more on usage see the :ref:`User Guide <univariate_feature_selection>`.
.. versionadded:: 1.0
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
The data matrix.
y : array-like of shape (n_samples,)
The target vector.
center : bool, default=True
Whether or not to center the data matrix `X` and the target vector `y`.
By default, `X` and `y` will be centered.
force_finite : bool, default=True
Whether or not to force the Pearson's R correlation to be finite.
In the particular case where some features in `X` or the target `y`
are constant, the Pearson's R correlation is not defined. When
`force_finite=False`, a correlation of `np.nan` is returned to
acknowledge this case. When `force_finite=True`, this value will be
forced to a minimal correlation of `0.0`.
.. versionadded:: 1.1
Returns
-------
correlation_coefficient : ndarray of shape (n_features,)
Pearson's R correlation coefficients of features.
See Also
--------
f_regression: Univariate linear regression tests returning f-statistic
and p-values.
mutual_info_regression: Mutual information for a continuous target.
f_classif: ANOVA F-value between label/feature for classification tasks.
chi2: Chi-squared stats of non-negative features for classification tasks.
Examples
--------
>>> from sklearn.datasets import make_regression
>>> from sklearn.feature_selection import r_regression
>>> X, y = make_regression(
... n_samples=50, n_features=3, n_informative=1, noise=1e-4, random_state=42
... )
>>> r_regression(X, y)
array([-0.15..., 1. , -0.22...])
"""
X, y = check_X_y(X, y, accept_sparse=["csr", "csc", "coo"], dtype=np.float64)
n_samples = X.shape[0]
# Compute centered values
# Note that E[(x - mean(x))*(y - mean(y))] = E[x*(y - mean(y))], so we
# need not center X
if center:
y = y - np.mean(y)
# TODO: for Scipy <= 1.10, `isspmatrix(X)` returns `True` for sparse arrays.
# Here, we check the output of the `.mean` operation that returns a `np.matrix`
# for sparse matrices while a `np.array` for dense and sparse arrays.
# We can reconsider using `isspmatrix` when the minimum version is
# SciPy >= 1.11
X_means = X.mean(axis=0)
X_means = X_means.getA1() if isinstance(X_means, np.matrix) else X_means
# Compute the scaled standard deviations via moments
X_norms = np.sqrt(row_norms(X.T, squared=True) - n_samples * X_means**2)
else:
X_norms = row_norms(X.T)
correlation_coefficient = safe_sparse_dot(y, X)
with np.errstate(divide="ignore", invalid="ignore"):
correlation_coefficient /= X_norms
correlation_coefficient /= np.linalg.norm(y)
if force_finite and not np.isfinite(correlation_coefficient).all():
# case where the target or some features are constant
# the correlation coefficient(s) is/are set to the minimum (i.e. 0.0)
nan_mask = np.isnan(correlation_coefficient)
correlation_coefficient[nan_mask] = 0.0
return correlation_coefficient
@validate_params(
{
"X": ["array-like", "sparse matrix"],
"y": ["array-like"],
"center": ["boolean"],
"force_finite": ["boolean"],
},
prefer_skip_nested_validation=True,
)
def f_regression(X, y, *, center=True, force_finite=True):
"""Univariate linear regression tests returning F-statistic and p-values.
Quick linear model for testing the effect of a single regressor,
sequentially for many regressors.
This is done in 2 steps:
1. The cross correlation between each regressor and the target is computed
using :func:`r_regression` as::
E[(X[:, i] - mean(X[:, i])) * (y - mean(y))] / (std(X[:, i]) * std(y))
2. It is converted to an F score and then to a p-value.
:func:`f_regression` is derived from :func:`r_regression` and will rank
features in the same order if all the features are positively correlated
with the target.
Note however that contrary to :func:`f_regression`, :func:`r_regression`
values lie in [-1, 1] and can thus be negative. :func:`f_regression` is
therefore recommended as a feature selection criterion to identify
potentially predictive feature for a downstream classifier, irrespective of
the sign of the association with the target variable.
Furthermore :func:`f_regression` returns p-values while
:func:`r_regression` does not.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
X : {array-like, sparse matrix} of shape (n_samples, n_features)
The data matrix.
y : array-like of shape (n_samples,)
The target vector.
center : bool, default=True
Whether or not to center the data matrix `X` and the target vector `y`.
By default, `X` and `y` will be centered.
force_finite : bool, default=True
Whether or not to force the F-statistics and associated p-values to
be finite. There are two cases where the F-statistic is expected to not
be finite:
- when the target `y` or some features in `X` are constant. In this
case, the Pearson's R correlation is not defined leading to obtain
`np.nan` values in the F-statistic and p-value. When
`force_finite=True`, the F-statistic is set to `0.0` and the
associated p-value is set to `1.0`.
- when a feature in `X` is perfectly correlated (or
anti-correlated) with the target `y`. In this case, the F-statistic
is expected to be `np.inf`. When `force_finite=True`, the F-statistic
is set to `np.finfo(dtype).max` and the associated p-value is set to
`0.0`.
.. versionadded:: 1.1
Returns
-------
f_statistic : ndarray of shape (n_features,)
F-statistic for each feature.
p_values : ndarray of shape (n_features,)
P-values associated with the F-statistic.
See Also
--------
r_regression: Pearson's R between label/feature for regression tasks.
f_classif: ANOVA F-value between label/feature for classification tasks.
chi2: Chi-squared stats of non-negative features for classification tasks.
SelectKBest: Select features based on the k highest scores.
SelectFpr: Select features based on a false positive rate test.
SelectFdr: Select features based on an estimated false discovery rate.
SelectFwe: Select features based on family-wise error rate.
SelectPercentile: Select features based on percentile of the highest
scores.
Examples
--------
>>> from sklearn.datasets import make_regression
>>> from sklearn.feature_selection import f_regression
>>> X, y = make_regression(
... n_samples=50, n_features=3, n_informative=1, noise=1e-4, random_state=42
... )
>>> f_statistic, p_values = f_regression(X, y)
>>> f_statistic
array([1.2...+00, 2.6...+13, 2.6...+00])
>>> p_values
array([2.7..., 1.5..., 1.0...])
"""
correlation_coefficient = r_regression(
X, y, center=center, force_finite=force_finite
)
deg_of_freedom = y.size - (2 if center else 1)
corr_coef_squared = correlation_coefficient**2
with np.errstate(divide="ignore", invalid="ignore"):
f_statistic = corr_coef_squared / (1 - corr_coef_squared) * deg_of_freedom
p_values = stats.f.sf(f_statistic, 1, deg_of_freedom)
if force_finite and not np.isfinite(f_statistic).all():
# case where there is a perfect (anti-)correlation
# f-statistics can be set to the maximum and p-values to zero
mask_inf = np.isinf(f_statistic)
f_statistic[mask_inf] = np.finfo(f_statistic.dtype).max
# case where the target or some features are constant
# f-statistics would be minimum and thus p-values large
mask_nan = np.isnan(f_statistic)
f_statistic[mask_nan] = 0.0
p_values[mask_nan] = 1.0
return f_statistic, p_values
######################################################################
# Base classes
class _BaseFilter(SelectorMixin, BaseEstimator):
"""Initialize the univariate feature selection.
Parameters
----------
score_func : callable
Function taking two arrays X and y, and returning a pair of arrays
(scores, pvalues) or a single array with scores.
"""
_parameter_constraints: dict = {"score_func": [callable]}
def __init__(self, score_func):
self.score_func = score_func
@_fit_context(prefer_skip_nested_validation=True)
def fit(self, X, y=None):
"""Run score function on (X, y) and get the appropriate features.
Parameters
----------
X : array-like of shape (n_samples, n_features)
The training input samples.
y : array-like of shape (n_samples,) or None
The target values (class labels in classification, real numbers in
regression). If the selector is unsupervised then `y` can be set to `None`.
Returns
-------
self : object
Returns the instance itself.
"""
if y is None:
X = self._validate_data(X, accept_sparse=["csr", "csc"])
else:
X, y = self._validate_data(
X, y, accept_sparse=["csr", "csc"], multi_output=True
)
self._check_params(X, y)
score_func_ret = self.score_func(X, y)
if isinstance(score_func_ret, (list, tuple)):
self.scores_, self.pvalues_ = score_func_ret
self.pvalues_ = np.asarray(self.pvalues_)
else:
self.scores_ = score_func_ret
self.pvalues_ = None
self.scores_ = np.asarray(self.scores_)
return self
def _check_params(self, X, y):
pass
def _more_tags(self):
return {"requires_y": True}
######################################################################
# Specific filters
######################################################################
class SelectPercentile(_BaseFilter):
"""Select features according to a percentile of the highest scores.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
score_func : callable, default=f_classif
Function taking two arrays X and y, and returning a pair of arrays
(scores, pvalues) or a single array with scores.
Default is f_classif (see below "See Also"). The default function only
works with classification tasks.
.. versionadded:: 0.18
percentile : int, default=10
Percent of features to keep.
Attributes
----------
scores_ : array-like of shape (n_features,)
Scores of features.
pvalues_ : array-like of shape (n_features,)
p-values of feature scores, None if `score_func` returned only scores.
n_features_in_ : int
Number of features seen during :term:`fit`.
.. versionadded:: 0.24
feature_names_in_ : ndarray of shape (`n_features_in_`,)
Names of features seen during :term:`fit`. Defined only when `X`
has feature names that are all strings.
.. versionadded:: 1.0
See Also
--------
f_classif : ANOVA F-value between label/feature for classification tasks.
mutual_info_classif : Mutual information for a discrete target.
chi2 : Chi-squared stats of non-negative features for classification tasks.
f_regression : F-value between label/feature for regression tasks.
mutual_info_regression : Mutual information for a continuous target.
SelectKBest : Select features based on the k highest scores.
SelectFpr : Select features based on a false positive rate test.
SelectFdr : Select features based on an estimated false discovery rate.
SelectFwe : Select features based on family-wise error rate.
GenericUnivariateSelect : Univariate feature selector with configurable
mode.
Notes
-----
Ties between features with equal scores will be broken in an unspecified
way.
This filter supports unsupervised feature selection that only requests `X` for
computing the scores.
Examples
--------
>>> from sklearn.datasets import load_digits
>>> from sklearn.feature_selection import SelectPercentile, chi2
>>> X, y = load_digits(return_X_y=True)
>>> X.shape
(1797, 64)
>>> X_new = SelectPercentile(chi2, percentile=10).fit_transform(X, y)
>>> X_new.shape
(1797, 7)
"""
_parameter_constraints: dict = {
**_BaseFilter._parameter_constraints,
"percentile": [Interval(Real, 0, 100, closed="both")],
}
def __init__(self, score_func=f_classif, *, percentile=10):
super().__init__(score_func=score_func)
self.percentile = percentile
def _get_support_mask(self):
check_is_fitted(self)
# Cater for NaNs
if self.percentile == 100:
return np.ones(len(self.scores_), dtype=bool)
elif self.percentile == 0:
return np.zeros(len(self.scores_), dtype=bool)
scores = _clean_nans(self.scores_)
threshold = np.percentile(scores, 100 - self.percentile)
mask = scores > threshold
ties = np.where(scores == threshold)[0]
if len(ties):
max_feats = int(len(scores) * self.percentile / 100)
kept_ties = ties[: max_feats - mask.sum()]
mask[kept_ties] = True
return mask
def _more_tags(self):
return {"requires_y": False}
class SelectKBest(_BaseFilter):
"""Select features according to the k highest scores.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
score_func : callable, default=f_classif
Function taking two arrays X and y, and returning a pair of arrays
(scores, pvalues) or a single array with scores.
Default is f_classif (see below "See Also"). The default function only
works with classification tasks.
.. versionadded:: 0.18
k : int or "all", default=10
Number of top features to select.
The "all" option bypasses selection, for use in a parameter search.
Attributes
----------
scores_ : array-like of shape (n_features,)
Scores of features.
pvalues_ : array-like of shape (n_features,)
p-values of feature scores, None if `score_func` returned only scores.
n_features_in_ : int
Number of features seen during :term:`fit`.
.. versionadded:: 0.24
feature_names_in_ : ndarray of shape (`n_features_in_`,)
Names of features seen during :term:`fit`. Defined only when `X`
has feature names that are all strings.
.. versionadded:: 1.0
See Also
--------
f_classif: ANOVA F-value between label/feature for classification tasks.
mutual_info_classif: Mutual information for a discrete target.
chi2: Chi-squared stats of non-negative features for classification tasks.
f_regression: F-value between label/feature for regression tasks.
mutual_info_regression: Mutual information for a continuous target.
SelectPercentile: Select features based on percentile of the highest
scores.
SelectFpr : Select features based on a false positive rate test.
SelectFdr : Select features based on an estimated false discovery rate.
SelectFwe : Select features based on family-wise error rate.
GenericUnivariateSelect : Univariate feature selector with configurable
mode.
Notes
-----
Ties between features with equal scores will be broken in an unspecified
way.
This filter supports unsupervised feature selection that only requests `X` for
computing the scores.
Examples
--------
>>> from sklearn.datasets import load_digits
>>> from sklearn.feature_selection import SelectKBest, chi2
>>> X, y = load_digits(return_X_y=True)
>>> X.shape
(1797, 64)
>>> X_new = SelectKBest(chi2, k=20).fit_transform(X, y)
>>> X_new.shape
(1797, 20)
"""
_parameter_constraints: dict = {
**_BaseFilter._parameter_constraints,
"k": [StrOptions({"all"}), Interval(Integral, 0, None, closed="left")],
}
def __init__(self, score_func=f_classif, *, k=10):
super().__init__(score_func=score_func)
self.k = k
def _check_params(self, X, y):
if not isinstance(self.k, str) and self.k > X.shape[1]:
warnings.warn(
f"k={self.k} is greater than n_features={X.shape[1]}. "
"All the features will be returned."
)
def _get_support_mask(self):
check_is_fitted(self)
if self.k == "all":
return np.ones(self.scores_.shape, dtype=bool)
elif self.k == 0:
return np.zeros(self.scores_.shape, dtype=bool)
else:
scores = _clean_nans(self.scores_)
mask = np.zeros(scores.shape, dtype=bool)
# Request a stable sort. Mergesort takes more memory (~40MB per
# megafeature on x86-64).
mask[np.argsort(scores, kind="mergesort")[-self.k :]] = 1
return mask
def _more_tags(self):
return {"requires_y": False}
class SelectFpr(_BaseFilter):
"""Filter: Select the pvalues below alpha based on a FPR test.
FPR test stands for False Positive Rate test. It controls the total
amount of false detections.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
score_func : callable, default=f_classif
Function taking two arrays X and y, and returning a pair of arrays
(scores, pvalues).
Default is f_classif (see below "See Also"). The default function only
works with classification tasks.
alpha : float, default=5e-2
Features with p-values less than `alpha` are selected.
Attributes
----------
scores_ : array-like of shape (n_features,)
Scores of features.
pvalues_ : array-like of shape (n_features,)
p-values of feature scores.
n_features_in_ : int
Number of features seen during :term:`fit`.
.. versionadded:: 0.24
feature_names_in_ : ndarray of shape (`n_features_in_`,)
Names of features seen during :term:`fit`. Defined only when `X`
has feature names that are all strings.
.. versionadded:: 1.0
See Also
--------
f_classif : ANOVA F-value between label/feature for classification tasks.
chi2 : Chi-squared stats of non-negative features for classification tasks.
mutual_info_classif: Mutual information for a discrete target.
f_regression : F-value between label/feature for regression tasks.
mutual_info_regression : Mutual information for a continuous target.
SelectPercentile : Select features based on percentile of the highest
scores.
SelectKBest : Select features based on the k highest scores.
SelectFdr : Select features based on an estimated false discovery rate.
SelectFwe : Select features based on family-wise error rate.
GenericUnivariateSelect : Univariate feature selector with configurable
mode.
Examples
--------
>>> from sklearn.datasets import load_breast_cancer
>>> from sklearn.feature_selection import SelectFpr, chi2
>>> X, y = load_breast_cancer(return_X_y=True)
>>> X.shape
(569, 30)
>>> X_new = SelectFpr(chi2, alpha=0.01).fit_transform(X, y)
>>> X_new.shape
(569, 16)
"""
_parameter_constraints: dict = {
**_BaseFilter._parameter_constraints,
"alpha": [Interval(Real, 0, 1, closed="both")],
}
def __init__(self, score_func=f_classif, *, alpha=5e-2):
super().__init__(score_func=score_func)
self.alpha = alpha
def _get_support_mask(self):
check_is_fitted(self)
return self.pvalues_ < self.alpha
class SelectFdr(_BaseFilter):
"""Filter: Select the p-values for an estimated false discovery rate.
This uses the Benjamini-Hochberg procedure. ``alpha`` is an upper bound
on the expected false discovery rate.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
score_func : callable, default=f_classif
Function taking two arrays X and y, and returning a pair of arrays
(scores, pvalues).
Default is f_classif (see below "See Also"). The default function only
works with classification tasks.
alpha : float, default=5e-2
The highest uncorrected p-value for features to keep.
Attributes
----------
scores_ : array-like of shape (n_features,)
Scores of features.
pvalues_ : array-like of shape (n_features,)
p-values of feature scores.
n_features_in_ : int
Number of features seen during :term:`fit`.
.. versionadded:: 0.24
feature_names_in_ : ndarray of shape (`n_features_in_`,)
Names of features seen during :term:`fit`. Defined only when `X`
has feature names that are all strings.
.. versionadded:: 1.0
See Also
--------
f_classif : ANOVA F-value between label/feature for classification tasks.
mutual_info_classif : Mutual information for a discrete target.
chi2 : Chi-squared stats of non-negative features for classification tasks.
f_regression : F-value between label/feature for regression tasks.
mutual_info_regression : Mutual information for a continuous target.
SelectPercentile : Select features based on percentile of the highest
scores.
SelectKBest : Select features based on the k highest scores.
SelectFpr : Select features based on a false positive rate test.
SelectFwe : Select features based on family-wise error rate.
GenericUnivariateSelect : Univariate feature selector with configurable
mode.
References
----------
https://en.wikipedia.org/wiki/False_discovery_rate
Examples
--------
>>> from sklearn.datasets import load_breast_cancer
>>> from sklearn.feature_selection import SelectFdr, chi2
>>> X, y = load_breast_cancer(return_X_y=True)
>>> X.shape
(569, 30)
>>> X_new = SelectFdr(chi2, alpha=0.01).fit_transform(X, y)
>>> X_new.shape
(569, 16)
"""
_parameter_constraints: dict = {
**_BaseFilter._parameter_constraints,
"alpha": [Interval(Real, 0, 1, closed="both")],
}
def __init__(self, score_func=f_classif, *, alpha=5e-2):
super().__init__(score_func=score_func)
self.alpha = alpha
def _get_support_mask(self):
check_is_fitted(self)
n_features = len(self.pvalues_)
sv = np.sort(self.pvalues_)
selected = sv[
sv <= float(self.alpha) / n_features * np.arange(1, n_features + 1)
]
if selected.size == 0:
return np.zeros_like(self.pvalues_, dtype=bool)
return self.pvalues_ <= selected.max()
class SelectFwe(_BaseFilter):
"""Filter: Select the p-values corresponding to Family-wise error rate.
Read more in the :ref:`User Guide <univariate_feature_selection>`.
Parameters
----------
score_func : callable, default=f_classif
Function taking two arrays X and y, and returning a pair of arrays
(scores, pvalues).
Default is f_classif (see below "See Also"). The default function only
works with classification tasks.
alpha : float, default=5e-2
The highest uncorrected p-value for features to keep.
Attributes
----------
scores_ : array-like of shape (n_features,)
Scores of features.
pvalues_ : array-like of shape (n_features,)
p-values of feature scores.
n_features_in_ : int
Number of features seen during :term:`fit`.
.. versionadded:: 0.24