-
-
Notifications
You must be signed in to change notification settings - Fork 26.9k
Expand file tree
/
Copy path_testing.py
More file actions
1506 lines (1289 loc) · 51.8 KB
/
_testing.py
File metadata and controls
1506 lines (1289 loc) · 51.8 KB
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
"""Testing utilities."""
# Authors: The scikit-learn developers
# SPDX-License-Identifier: BSD-3-Clause
import atexit
import contextlib
import functools
import importlib
import inspect
import os
import os.path as op
import re
import shutil
import sys
import tempfile
import textwrap
import unittest
import warnings
from collections import defaultdict, namedtuple
from collections.abc import Iterable
from dataclasses import dataclass
from difflib import context_diff
from functools import wraps
from inspect import signature
from itertools import chain, groupby
from subprocess import STDOUT, CalledProcessError, TimeoutExpired, check_output
import joblib
import numpy as np
import scipy as sp
from numpy.testing import assert_allclose as np_assert_allclose
from numpy.testing import (
assert_almost_equal,
assert_array_almost_equal,
assert_array_equal,
assert_array_less,
)
from sklearn import __file__ as sklearn_path
from sklearn.utils import (
ClassifierTags,
RegressorTags,
Tags,
TargetTags,
TransformerTags,
)
from sklearn.utils._array_api import _check_array_api_dispatch
from sklearn.utils.fixes import (
_IS_32BIT,
VisibleDeprecationWarning,
_in_unstable_openblas_configuration,
)
from sklearn.utils.multiclass import check_classification_targets
from sklearn.utils.validation import check_array, check_is_fitted, check_X_y
__all__ = [
"SkipTest",
"assert_allclose",
"assert_almost_equal",
"assert_array_almost_equal",
"assert_array_equal",
"assert_array_less",
"assert_run_python_script_without_output",
]
SkipTest = unittest.case.SkipTest
def ignore_warnings(obj=None, category=Warning):
"""Context manager and decorator to ignore warnings.
Note: Using this (in both variants) will clear all warnings
from all python modules loaded. In case you need to test
cross-module-warning-logging, this is not your tool of choice.
Parameters
----------
obj : callable, default=None
callable where you want to ignore the warnings.
category : warning class, default=Warning
The category to filter. If Warning, all categories will be muted.
Examples
--------
>>> import warnings
>>> from sklearn.utils._testing import ignore_warnings
>>> with ignore_warnings():
... warnings.warn('buhuhuhu')
>>> def nasty_warn():
... warnings.warn('buhuhuhu')
... print(42)
>>> ignore_warnings(nasty_warn)()
42
"""
if isinstance(obj, type) and issubclass(obj, Warning):
# Avoid common pitfall of passing category as the first positional
# argument which result in the test not being run
warning_name = obj.__name__
raise ValueError(
"'obj' should be a callable where you want to ignore warnings. "
"You passed a warning class instead: 'obj={warning_name}'. "
"If you want to pass a warning class to ignore_warnings, "
"you should use 'category={warning_name}'".format(warning_name=warning_name)
)
elif callable(obj):
return _IgnoreWarnings(category=category)(obj)
else:
return _IgnoreWarnings(category=category)
class _IgnoreWarnings:
"""Improved and simplified Python warnings context manager and decorator.
This class allows the user to ignore the warnings raised by a function.
Copied from Python 2.7.5 and modified as required.
Parameters
----------
category : tuple of warning class, default=Warning
The category to filter. By default, all the categories will be muted.
"""
def __init__(self, category):
self._record = True
self._module = sys.modules["warnings"]
self._entered = False
self.log = []
self.category = category
def __call__(self, fn):
"""Decorator to catch and hide warnings without visual nesting."""
@wraps(fn)
def wrapper(*args, **kwargs):
with warnings.catch_warnings():
warnings.simplefilter("ignore", self.category)
return fn(*args, **kwargs)
return wrapper
def __repr__(self):
args = []
if self._record:
args.append("record=True")
if self._module is not sys.modules["warnings"]:
args.append("module=%r" % self._module)
name = type(self).__name__
return "%s(%s)" % (name, ", ".join(args))
def __enter__(self):
if self._entered:
raise RuntimeError("Cannot enter %r twice" % self)
self._entered = True
self._filters = self._module.filters
self._module.filters = self._filters[:]
self._showwarning = self._module.showwarning
warnings.simplefilter("ignore", self.category)
def __exit__(self, *exc_info):
if not self._entered:
raise RuntimeError("Cannot exit %r without entering first" % self)
self._module.filters = self._filters
self._module.showwarning = self._showwarning
self.log[:] = []
def assert_allclose(
actual, desired, rtol=None, atol=0.0, equal_nan=True, err_msg="", verbose=True
):
"""dtype-aware variant of numpy.testing.assert_allclose
This variant introspects the least precise floating point dtype
in the input argument and automatically sets the relative tolerance
parameter to 1e-4 float32 and use 1e-7 otherwise (typically float64
in scikit-learn).
`atol` is always left to 0. by default. It should be adjusted manually
to an assertion-specific value in case there are null values expected
in `desired`.
The aggregate tolerance is `atol + rtol * abs(desired)`.
Parameters
----------
actual : array_like
Array obtained.
desired : array_like
Array desired.
rtol : float, optional, default=None
Relative tolerance.
If None, it is set based on the provided arrays' dtypes.
atol : float, optional, default=0.
Absolute tolerance.
equal_nan : bool, optional, default=True
If True, NaNs will compare equal.
err_msg : str, optional, default=''
The error message to be printed in case of failure.
verbose : bool, optional, default=True
If True, the conflicting values are appended to the error message.
Raises
------
AssertionError
If actual and desired are not equal up to specified precision.
See Also
--------
numpy.testing.assert_allclose
Examples
--------
>>> import numpy as np
>>> from sklearn.utils._testing import assert_allclose
>>> x = [1e-5, 1e-3, 1e-1]
>>> y = np.arccos(np.cos(x))
>>> assert_allclose(x, y, rtol=1e-5, atol=0)
>>> a = np.full(shape=10, fill_value=1e-5, dtype=np.float32)
>>> assert_allclose(a, 1e-5)
"""
dtypes = []
actual, desired = np.asanyarray(actual), np.asanyarray(desired)
dtypes = [actual.dtype, desired.dtype]
if rtol is None:
rtols = [1e-4 if dtype == np.float32 else 1e-7 for dtype in dtypes]
rtol = max(rtols)
np_assert_allclose(
actual,
desired,
rtol=rtol,
atol=atol,
equal_nan=equal_nan,
err_msg=err_msg,
verbose=verbose,
)
def assert_allclose_dense_sparse(x, y, rtol=1e-07, atol=1e-9, err_msg=""):
"""Assert allclose for sparse and dense data.
Both x and y need to be either sparse or dense, they
can't be mixed.
Parameters
----------
x : {array-like, sparse matrix}
First array to compare.
y : {array-like, sparse matrix}
Second array to compare.
rtol : float, default=1e-07
relative tolerance; see numpy.allclose.
atol : float, default=1e-9
absolute tolerance; see numpy.allclose. Note that the default here is
more tolerant than the default for numpy.testing.assert_allclose, where
atol=0.
err_msg : str, default=''
Error message to raise.
"""
if sp.sparse.issparse(x) and sp.sparse.issparse(y):
x = x.tocsr()
y = y.tocsr()
x.sum_duplicates()
y.sum_duplicates()
assert_array_equal(x.indices, y.indices, err_msg=err_msg)
assert_array_equal(x.indptr, y.indptr, err_msg=err_msg)
assert_allclose(x.data, y.data, rtol=rtol, atol=atol, err_msg=err_msg)
elif not sp.sparse.issparse(x) and not sp.sparse.issparse(y):
# both dense
assert_allclose(x, y, rtol=rtol, atol=atol, err_msg=err_msg)
else:
raise ValueError(
"Can only compare two sparse matrices, not a sparse matrix and an array."
)
def set_random_state(estimator, random_state=0):
"""Set random state of an estimator if it has the `random_state` param.
Parameters
----------
estimator : object
The estimator.
random_state : int, RandomState instance or None, default=0
Pseudo random number generator state.
Pass an int for reproducible results across multiple function calls.
See :term:`Glossary <random_state>`.
"""
if "random_state" in estimator.get_params():
estimator.set_params(random_state=random_state)
def _is_numpydoc():
try:
import numpydoc # noqa: F401
except (ImportError, AssertionError):
return False
else:
return True
try:
_check_array_api_dispatch(True)
ARRAY_API_COMPAT_FUNCTIONAL = True
except (ImportError, RuntimeError):
ARRAY_API_COMPAT_FUNCTIONAL = False
try:
import pytest
skip_if_32bit = pytest.mark.skipif(_IS_32BIT, reason="skipped on 32bit platforms")
fails_if_unstable_openblas = pytest.mark.xfail(
_in_unstable_openblas_configuration(),
reason="OpenBLAS is unstable for this configuration",
)
skip_if_no_parallel = pytest.mark.skipif(
not joblib.parallel.mp, reason="joblib is in serial mode"
)
skip_if_array_api_compat_not_configured = pytest.mark.skipif(
not ARRAY_API_COMPAT_FUNCTIONAL,
reason="SCIPY_ARRAY_API not set, or versions of NumPy/SciPy too old.",
)
# Decorator for tests involving both BLAS calls and multiprocessing.
#
# Under POSIX (e.g. Linux or OSX), using multiprocessing in conjunction
# with some implementation of BLAS (or other libraries that manage an
# internal posix thread pool) can cause a crash or a freeze of the Python
# process.
#
# In practice all known packaged distributions (from Linux distros or
# Anaconda) of BLAS under Linux seems to be safe. So we this problem seems
# to only impact OSX users.
#
# This wrapper makes it possible to skip tests that can possibly cause
# this crash under OS X with.
#
# Under Python 3.4+ it is possible to use the `forkserver` start method
# for multiprocessing to avoid this issue. However it can cause pickling
# errors on interactively defined functions. It therefore not enabled by
# default.
if_safe_multiprocessing_with_blas = pytest.mark.skipif(
sys.platform == "darwin", reason="Possible multi-process bug with some BLAS"
)
skip_if_no_numpydoc = pytest.mark.skipif(
not _is_numpydoc(),
reason="numpydoc is required to test the docstrings",
)
except ImportError:
pass
def check_skip_network():
if int(os.environ.get("SKLEARN_SKIP_NETWORK_TESTS", 0)):
raise SkipTest("Text tutorial requires large dataset download")
def _delete_folder(folder_path, warn=False):
"""Utility function to cleanup a temporary folder if still existing.
Copy from joblib.pool (for independence).
"""
try:
if os.path.exists(folder_path):
# This can fail under windows,
# but will succeed when called by atexit
shutil.rmtree(folder_path)
except OSError:
if warn:
warnings.warn("Could not delete temporary folder %s" % folder_path)
class TempMemmap:
"""
Parameters
----------
data
mmap_mode : str, default='r'
"""
def __init__(self, data, mmap_mode="r"):
self.mmap_mode = mmap_mode
self.data = data
def __enter__(self):
data_read_only, self.temp_folder = create_memmap_backed_data(
self.data, mmap_mode=self.mmap_mode, return_folder=True
)
return data_read_only
def __exit__(self, exc_type, exc_val, exc_tb):
_delete_folder(self.temp_folder)
def create_memmap_backed_data(data, mmap_mode="r", return_folder=False):
"""
Parameters
----------
data
mmap_mode : str, default='r'
return_folder : bool, default=False
"""
temp_folder = tempfile.mkdtemp(prefix="sklearn_testing_")
atexit.register(functools.partial(_delete_folder, temp_folder, warn=True))
filename = op.join(temp_folder, "data.pkl")
joblib.dump(data, filename)
memmap_backed_data = joblib.load(filename, mmap_mode=mmap_mode)
result = (
memmap_backed_data if not return_folder else (memmap_backed_data, temp_folder)
)
return result
# Utils to test docstrings
def _get_args(function, varargs=False):
"""Helper to get function arguments."""
try:
params = signature(function).parameters
except ValueError:
# Error on builtin C function
return []
args = [
key
for key, param in params.items()
if param.kind not in (param.VAR_POSITIONAL, param.VAR_KEYWORD)
]
if varargs:
varargs = [
param.name
for param in params.values()
if param.kind == param.VAR_POSITIONAL
]
if len(varargs) == 0:
varargs = None
return args, varargs
else:
return args
def _get_func_name(func):
"""Get function full name.
Parameters
----------
func : callable
The function object.
Returns
-------
name : str
The function name.
"""
parts = []
module = inspect.getmodule(func)
if module:
parts.append(module.__name__)
qualname = func.__qualname__
if qualname != func.__name__:
parts.append(qualname[: qualname.find(".")])
parts.append(func.__name__)
return ".".join(parts)
def check_docstring_parameters(func, doc=None, ignore=None):
"""Helper to check docstring.
Parameters
----------
func : callable
The function object to test.
doc : str, default=None
Docstring if it is passed manually to the test.
ignore : list, default=None
Parameters to ignore.
Returns
-------
incorrect : list
A list of string describing the incorrect results.
"""
from numpydoc import docscrape
incorrect = []
ignore = [] if ignore is None else ignore
func_name = _get_func_name(func)
if not func_name.startswith("sklearn.") or func_name.startswith(
"sklearn.externals"
):
return incorrect
# Don't check docstring for property-functions
if inspect.isdatadescriptor(func):
return incorrect
# Don't check docstring for setup / teardown pytest functions
if func_name.split(".")[-1] in ("setup_module", "teardown_module"):
return incorrect
# Dont check estimator_checks module
if func_name.split(".")[2] == "estimator_checks":
return incorrect
# Get the arguments from the function signature
param_signature = list(filter(lambda x: x not in ignore, _get_args(func)))
# drop self
if len(param_signature) > 0 and param_signature[0] == "self":
param_signature.remove("self")
# Analyze function's docstring
if doc is None:
records = []
with warnings.catch_warnings(record=True):
warnings.simplefilter("error", UserWarning)
try:
doc = docscrape.FunctionDoc(func)
except UserWarning as exp:
if "potentially wrong underline length" in str(exp):
# Catch warning raised as of numpydoc 1.2 when
# the underline length for a section of a docstring
# is not consistent.
message = str(exp).split("\n")[:3]
incorrect += [f"In function: {func_name}"] + message
return incorrect
records.append(str(exp))
except Exception as exp:
incorrect += [func_name + " parsing error: " + str(exp)]
return incorrect
if len(records):
raise RuntimeError("Error for %s:\n%s" % (func_name, records[0]))
param_docs = []
for name, type_definition, param_doc in doc["Parameters"]:
# Type hints are empty only if parameter name ended with :
if not type_definition.strip():
if ":" in name and name[: name.index(":")][-1:].strip():
incorrect += [
func_name
+ " There was no space between the param name and colon (%r)" % name
]
elif name.rstrip().endswith(":"):
incorrect += [
func_name
+ " Parameter %r has an empty type spec. Remove the colon"
% (name.lstrip())
]
# Create a list of parameters to compare with the parameters gotten
# from the func signature
if "*" not in name:
param_docs.append(name.split(":")[0].strip("` "))
# If one of the docstring's parameters had an error then return that
# incorrect message
if len(incorrect) > 0:
return incorrect
# Remove the parameters that should be ignored from list
param_docs = list(filter(lambda x: x not in ignore, param_docs))
# The following is derived from pytest, Copyright (c) 2004-2017 Holger
# Krekel and others, Licensed under MIT License. See
# https://github.com/pytest-dev/pytest
message = []
for i in range(min(len(param_docs), len(param_signature))):
if param_signature[i] != param_docs[i]:
message += [
"There's a parameter name mismatch in function"
" docstring w.r.t. function signature, at index %s"
" diff: %r != %r" % (i, param_signature[i], param_docs[i])
]
break
if len(param_signature) > len(param_docs):
message += [
"Parameters in function docstring have less items w.r.t."
" function signature, first missing item: %s"
% param_signature[len(param_docs)]
]
elif len(param_signature) < len(param_docs):
message += [
"Parameters in function docstring have more items w.r.t."
" function signature, first extra item: %s"
% param_docs[len(param_signature)]
]
# If there wasn't any difference in the parameters themselves between
# docstring and signature including having the same length then return
# empty list
if len(message) == 0:
return []
import difflib
import pprint
param_docs_formatted = pprint.pformat(param_docs).splitlines()
param_signature_formatted = pprint.pformat(param_signature).splitlines()
message += ["Full diff:"]
message.extend(
line.strip()
for line in difflib.ndiff(param_signature_formatted, param_docs_formatted)
)
incorrect.extend(message)
# Prepend function name
incorrect = ["In function: " + func_name] + incorrect
return incorrect
def _check_item_included(item_name, args):
"""Helper to check if item should be included in checking."""
if args.include is not True and item_name not in args.include:
return False
if args.exclude is not None and item_name in args.exclude:
return False
return True
def _diff_key(line):
"""Key for grouping output from `context_diff`."""
if line.startswith(" "):
return " "
elif line.startswith("- "):
return "- "
elif line.startswith("+ "):
return "+ "
elif line.startswith("! "):
return "! "
return None
def _get_diff_msg(docstrings_grouped):
"""Get message showing the difference between type/desc docstrings of all objects.
`docstrings_grouped` keys should be the type/desc docstrings and values are a list
of objects with that docstring. Objects with the same type/desc docstring are
thus grouped together.
"""
msg_diff = ""
ref_str = ""
ref_group = []
for docstring, group in docstrings_grouped.items():
if not ref_str and not ref_group:
ref_str += docstring
ref_group.extend(group)
diff = list(
context_diff(
ref_str.split(),
docstring.split(),
fromfile=str(ref_group),
tofile=str(group),
n=8,
)
)
# Add header
msg_diff += "".join((diff[:3]))
# Group consecutive 'diff' words to shorten error message
for start, group in groupby(diff[3:], key=_diff_key):
if start is None:
msg_diff += "\n" + "\n".join(group)
else:
msg_diff += "\n" + start + " ".join(word[2:] for word in group)
# Add new lines at end of diff, to separate comparisons
msg_diff += "\n\n"
return msg_diff
def _check_consistency_items(
items_docs,
type_or_desc,
section,
n_objects,
descr_regex_pattern="",
ignore_types=tuple(),
):
"""Helper to check docstring consistency of all `items_docs`.
If item is not present in all objects, checking is skipped and warning raised.
If `regex` provided, match descriptions to all descriptions.
Parameters
----------
items_doc : dict of dict of str
Dictionary where the key is the string type or description, value is
a dictionary where the key is "type description" or "description"
and the value is a list of object names with the same string type or
description.
type_or_desc : {"type description", "description"}
Whether to check type description or description between objects.
section : {"Parameters", "Attributes", "Returns"}
Name of the section type.
n_objects : int
Total number of objects.
descr_regex_pattern : str, default=""
Regex pattern to match for description of all objects.
Ignored when `type_or_desc="type description".
ignore_types : tuple of str, default=()
Tuple of parameter/attribute/return names for which type description
matching is ignored. Ignored when `type_or_desc="description".
"""
skipped = []
for item_name, docstrings_grouped in items_docs.items():
# If item not found in all objects, skip
if sum([len(objs) for objs in docstrings_grouped.values()]) < n_objects:
skipped.append(item_name)
# If regex provided, match to all descriptions
elif type_or_desc == "description" and descr_regex_pattern:
not_matched = []
for docstring, group in docstrings_grouped.items():
if not re.search(descr_regex_pattern, docstring):
not_matched.extend(group)
if not_matched:
msg = textwrap.fill(
f"The description of {section[:-1]} '{item_name}' in {not_matched}"
f" does not match 'descr_regex_pattern': {descr_regex_pattern} "
)
raise AssertionError(msg)
# Skip type checking for items in `ignore_types`
elif type_or_desc == "type specification" and item_name in ignore_types:
continue
# Otherwise, if more than one key, docstrings not consistent between objects
elif len(docstrings_grouped.keys()) > 1:
msg_diff = _get_diff_msg(docstrings_grouped)
obj_groups = " and ".join(
str(group) for group in docstrings_grouped.values()
)
msg = textwrap.fill(
f"The {type_or_desc} of {section[:-1]} '{item_name}' is inconsistent "
f"between {obj_groups}:"
)
msg += msg_diff
raise AssertionError(msg)
if skipped:
warnings.warn(
f"Checking was skipped for {section}: {skipped} as they were "
"not found in all objects."
)
def assert_docstring_consistency(
objects,
include_params=False,
exclude_params=None,
include_attrs=False,
exclude_attrs=None,
include_returns=False,
exclude_returns=None,
descr_regex_pattern=None,
ignore_types=tuple(),
):
r"""Check consistency between docstring parameters/attributes/returns of objects.
Checks if parameters/attributes/returns have the same type specification and
description (ignoring whitespace) across `objects`. Intended to be used for
related classes/functions/data descriptors.
Entries that do not appear across all `objects` are ignored.
Parameters
----------
objects : list of {classes, functions, data descriptors}
Objects to check.
Objects may be classes, functions or data descriptors with docstrings that
can be parsed by numpydoc.
include_params : list of str or bool, default=False
List of parameters to be included. If True, all parameters are included,
if False, checking is skipped for parameters.
Can only be set if `exclude_params` is None.
exclude_params : list of str or None, default=None
List of parameters to be excluded. If None, no parameters are excluded.
Can only be set if `include_params` is True.
include_attrs : list of str or bool, default=False
List of attributes to be included. If True, all attributes are included,
if False, checking is skipped for attributes.
Can only be set if `exclude_attrs` is None.
exclude_attrs : list of str or None, default=None
List of attributes to be excluded. If None, no attributes are excluded.
Can only be set if `include_attrs` is True.
include_returns : list of str or bool, default=False
List of returns to be included. If True, all returns are included,
if False, checking is skipped for returns.
Can only be set if `exclude_returns` is None.
exclude_returns : list of str or None, default=None
List of returns to be excluded. If None, no returns are excluded.
Can only be set if `include_returns` is True.
descr_regex_pattern : str, default=None
Regular expression to match to all descriptions of included
parameters/attributes/returns. If None, will revert to default behavior
of comparing descriptions between objects.
ignore_types : tuple of str, default=tuple()
Tuple of parameter/attribute/return names to exclude from type description
matching between objects.
Examples
--------
>>> from sklearn.metrics import (accuracy_score, classification_report,
... mean_absolute_error, mean_squared_error, median_absolute_error)
>>> from sklearn.utils._testing import assert_docstring_consistency
... # doctest: +SKIP
>>> assert_docstring_consistency([mean_absolute_error, mean_squared_error],
... include_params=['y_true', 'y_pred', 'sample_weight']) # doctest: +SKIP
>>> assert_docstring_consistency([median_absolute_error, mean_squared_error],
... include_params=True) # doctest: +SKIP
>>> assert_docstring_consistency([accuracy_score, classification_report],
... include_params=["y_true"],
... descr_regex_pattern=r"Ground truth \(correct\) (labels|target values)")
... # doctest: +SKIP
"""
from numpydoc.docscrape import NumpyDocString
Args = namedtuple("args", ["include", "exclude", "arg_name"])
def _create_args(include, exclude, arg_name, section_name):
if exclude and include is not True:
raise TypeError(
f"The 'exclude_{arg_name}' argument can be set only when the "
f"'include_{arg_name}' argument is True."
)
if include is False:
return {}
return {section_name: Args(include, exclude, arg_name)}
section_args = {
**_create_args(include_params, exclude_params, "params", "Parameters"),
**_create_args(include_attrs, exclude_attrs, "attrs", "Attributes"),
**_create_args(include_returns, exclude_returns, "returns", "Returns"),
}
objects_doc = dict()
for obj in objects:
if (
inspect.isdatadescriptor(obj)
or inspect.isfunction(obj)
or inspect.isclass(obj)
):
objects_doc[obj.__name__] = NumpyDocString(inspect.getdoc(obj))
else:
raise TypeError(
"All 'objects' must be one of: function, class or descriptor, "
f"got a: {type(obj)}."
)
n_objects = len(objects)
for section, args in section_args.items():
type_items = defaultdict(lambda: defaultdict(list))
desc_items = defaultdict(lambda: defaultdict(list))
for obj_name, obj_doc in objects_doc.items():
for item_name, type_def, desc in obj_doc[section]:
if _check_item_included(item_name, args):
# Normalize white space
type_def = " ".join(type_def.strip().split())
desc = " ".join(chain.from_iterable(line.split() for line in desc))
# Use string type/desc as key, to group consistent objs together
type_items[item_name][type_def].append(obj_name)
desc_items[item_name][desc].append(obj_name)
_check_consistency_items(
type_items,
"type specification",
section,
n_objects,
ignore_types=ignore_types,
)
_check_consistency_items(
desc_items,
"description",
section,
n_objects,
descr_regex_pattern=descr_regex_pattern,
)
def assert_run_python_script_without_output(source_code, pattern=".+", timeout=60):
"""Utility to check assertions in an independent Python subprocess.
The script provided in the source code should return 0 and the stdtout +
stderr should not match the pattern `pattern`.
This is a port from cloudpickle https://github.com/cloudpipe/cloudpickle
Parameters
----------
source_code : str
The Python source code to execute.
pattern : str
Pattern that the stdout + stderr should not match. By default, unless
stdout + stderr are both empty, an error will be raised.
timeout : int, default=60
Time in seconds before timeout.
"""
fd, source_file = tempfile.mkstemp(suffix="_src_test_sklearn.py")
os.close(fd)
try:
with open(source_file, "wb") as f:
f.write(source_code.encode("utf-8"))
cmd = [sys.executable, source_file]
cwd = op.normpath(op.join(op.dirname(sklearn_path), ".."))
env = os.environ.copy()
try:
env["PYTHONPATH"] = os.pathsep.join([cwd, env["PYTHONPATH"]])
except KeyError:
env["PYTHONPATH"] = cwd
kwargs = {"cwd": cwd, "stderr": STDOUT, "env": env}
# If coverage is running, pass the config file to the subprocess
coverage_rc = os.environ.get("COVERAGE_PROCESS_START")
if coverage_rc:
kwargs["env"]["COVERAGE_PROCESS_START"] = coverage_rc
kwargs["timeout"] = timeout
try:
try:
out = check_output(cmd, **kwargs)
except CalledProcessError as e:
raise RuntimeError(
"script errored with output:\n%s" % e.output.decode("utf-8")
)
out = out.decode("utf-8")
if re.search(pattern, out):
if pattern == ".+":
expectation = "Expected no output"
else:
expectation = f"The output was not supposed to match {pattern!r}"
message = f"{expectation}, got the following output instead: {out!r}"
raise AssertionError(message)
except TimeoutExpired as e:
raise RuntimeError(
"script timeout, output so far:\n%s" % e.output.decode("utf-8")
)
finally:
os.unlink(source_file)
def _convert_container(
container,
constructor_name,
columns_name=None,
dtype=None,
minversion=None,
categorical_feature_names=None,
):
"""Convert a given container to a specific array-like with a dtype.
Parameters
----------
container : array-like
The container to convert.
constructor_name : {"list", "tuple", "array", "sparse", "dataframe", \
"pandas", "series", "index", "slice", "sparse_csr", "sparse_csc", \
"sparse_csr_array", "sparse_csc_array", "pyarrow", "polars", \
"polars_series"}
The type of the returned container.
columns_name : index or array-like, default=None
For pandas/polars container supporting `columns_names`, it will affect
specific names.
dtype : dtype, default=None
Force the dtype of the container. Does not apply to `"slice"`
container.
minversion : str, default=None
Minimum version for package to install.
categorical_feature_names : list of str, default=None
List of column names to cast to categorical dtype.
Returns
-------
converted_container
"""
if constructor_name == "list":
if dtype is None:
return list(container)