Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Commit fc0150d

Browse files
jnothmanogrisel
jnothman
authored and
ogrisel
committed
DOC/FIX Address @arjoly's comments
1 parent dcb2417 commit fc0150d

File tree

6 files changed

+41
-71
lines changed

6 files changed

+41
-71
lines changed

‎doc/developers/utilities.rst

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -244,7 +244,7 @@ Multiclass and multilabel utility function
244244
a classification output is in label indicator matrix format.
245245

246246
- :func:`multiclass.unique_labels`: Helper function to extract an ordered
247-
array of unique labels from a list of labels.
247+
array of unique labels from different formats of target.
248248

249249

250250
Helper Functions

‎doc/modules/model_evaluation.rst

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -259,7 +259,7 @@ where :math:`1(x)` is the `indicator function
259259
>>> accuracy_score(y_true, y_pred, normalize=False)
260260
2
261261

262-
In the multilabel case with binary label indicators:
262+
In the multilabel case with binary label indicators: ::
263263

264264
>>> accuracy_score(np.array([[0.0, 1.0], [1.0, 1.0]]), np.ones((2, 2)))
265265
0.5
@@ -424,7 +424,7 @@ score is equal to the classification accuracy.
424424
>>> jaccard_similarity_score(y_true, y_pred, normalize=False)
425425
2
426426

427-
In the multilabel case with binary label indicators:
427+
In the multilabel case with binary label indicators: ::
428428

429429
>>> jaccard_similarity_score(np.array([[0.0, 1.0], [1.0, 1.0]]), np.ones((2, 2)))
430430
0.75

‎examples/plot_multilabel.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -72,7 +72,7 @@ def plot_subfigure(X, Y, subplot, title, transform):
7272
pl.title(title)
7373

7474
zero_class = np.where(Y[:, 0])
75-
one_class = np.where(Y[:, 0])
75+
one_class = np.where(Y[:, 1])
7676
pl.scatter(X[:, 0], X[:, 1], s=40, c='gray')
7777
pl.scatter(X[zero_class, 0], X[zero_class, 1], s=160, edgecolors='b',
7878
facecolors='none', linewidths=2, label='Class 1')

‎sklearn/metrics/metrics.py

Lines changed: 32 additions & 57 deletions
Original file line numberDiff line numberDiff line change
@@ -314,7 +314,7 @@ def average_precision_score(y_true, y_score, average="macro",
314314
315315
----------
316316
y_true : array, shape = [n_samples] or [n_samples, n_classes]
317-
True binary labels in binary indicator format.
317+
True binary labels in binary label indicators.
318318
319319
y_score : array, shape = [n_samples] or [n_samples, n_classes]
320320
Target scores, can either be probability estimates of the positive
@@ -426,7 +426,7 @@ def _average_binary_score(binary_metric, y_true, y_score, average,
426426
Parameters
427427
----------
428428
y_true : array, shape = [n_samples] or [n_samples, n_classes]
429-
True binary labels in binary indicator format.
429+
True binary labels in binary label indicators.
430430
431431
y_score : array, shape = [n_samples] or [n_samples, n_classes]
432432
Target scores, can either be probability estimates of the positive
@@ -527,7 +527,7 @@ def roc_auc_score(y_true, y_score, average="macro", sample_weight=None):
527527
Parameters
528528
----------
529529
y_true : array, shape = [n_samples] or [n_samples, n_classes]
530-
True binary labels in binary indicator format.
530+
True binary labels in binary label indicators.
531531
532532
y_score : array, shape = [n_samples] or [n_samples, n_classes]
533533
Target scores, can either be probability estimates of the positive
@@ -986,10 +986,10 @@ def zero_one_loss(y_true, y_pred, normalize=True, sample_weight=None):
986986
987987
Parameters
988988
----------
989-
y_true : array-like or list of labels or label indicator matrix
989+
y_true : array-like or label indicator matrix
990990
Ground truth (correct) labels.
991991
992-
y_pred : array-like or list of labels or label indicator matrix
992+
y_pred : array-like or label indicator matrix
993993
Predicted labels, as returned by a classifier.
994994
995995
normalize : bool, optional (default=True)
@@ -1025,17 +1025,10 @@ def zero_one_loss(y_true, y_pred, normalize=True, sample_weight=None):
10251025
>>> zero_one_loss(y_true, y_pred, normalize=False)
10261026
1
10271027
1028-
In the multilabel case with binary indicator format:
1028+
In the multilabel case with binary label indicators:
10291029
1030-
>>> zero_one_loss(np.array([[0.0, 1.0], [1.0, 1.0]]), np.ones((2, 2)))
1030+
>>> zero_one_loss(np.array([[0, 1], [1, 1]]), np.ones((2, 2)))
10311031
0.5
1032-
1033-
and with a list of labels format:
1034-
1035-
>>> zero_one_loss([(1, ), (3, )], [(1, 2), tuple()])
1036-
1.0
1037-
1038-
10391032
"""
10401033
score = accuracy_score(y_true, y_pred,
10411034
normalize=normalize,
@@ -1064,7 +1057,7 @@ def log_loss(y_true, y_pred, eps=1e-15, normalize=True):
10641057
10651058
Parameters
10661059
----------
1067-
y_true : array-like or list of labels or label indicator matrix
1060+
y_true : array-like or label indicator matrix
10681061
Ground truth (correct) labels for n_samples samples.
10691062
10701063
y_pred : array-like of float, shape = (n_samples, n_classes)
@@ -1139,10 +1132,10 @@ def jaccard_similarity_score(y_true, y_pred, normalize=True):
11391132
11401133
Parameters
11411134
----------
1142-
y_true : array-like or list of labels or label indicator matrix
1135+
y_true : array-like or label indicator matrix
11431136
Ground truth (correct) labels.
11441137
1145-
y_pred : array-like or list of labels or label indicator matrix
1138+
y_pred : array-like or label indicator matrix
11461139
Predicted labels, as returned by a classifier.
11471140
11481141
normalize : bool, optional (default=True)
@@ -1187,17 +1180,11 @@ def jaccard_similarity_score(y_true, y_pred, normalize=True):
11871180
>>> jaccard_similarity_score(y_true, y_pred, normalize=False)
11881181
2
11891182
1190-
In the multilabel case with binary indicator format:
1183+
In the multilabel case with binary label indicators:
11911184
1192-
>>> jaccard_similarity_score(np.array([[0.0, 1.0], [1.0, 1.0]]),\
1185+
>>> jaccard_similarity_score(np.array([[0, 1], [1, 1]]),\
11931186
np.ones((2, 2)))
11941187
0.75
1195-
1196-
and with a list of labels format:
1197-
1198-
>>> jaccard_similarity_score([(1, ), (3, )], [(1, 2), tuple()])
1199-
0.25
1200-
12011188
"""
12021189

12031190
# Compute accuracy for each possible representation
@@ -1252,10 +1239,10 @@ def accuracy_score(y_true, y_pred, normalize=True, sample_weight=None):
12521239
12531240
Parameters
12541241
----------
1255-
y_true : array-like or list of labels or label indicator matrix
1242+
y_true : array-like or label indicator matrix
12561243
Ground truth (correct) labels.
12571244
1258-
y_pred : array-like or list of labels or label indicator matrix
1245+
y_pred : array-like or label indicator matrix
12591246
Predicted labels, as returned by a classifier.
12601247
12611248
normalize : bool, optional (default=True)
@@ -1295,16 +1282,10 @@ def accuracy_score(y_true, y_pred, normalize=True, sample_weight=None):
12951282
>>> accuracy_score(y_true, y_pred, normalize=False)
12961283
2
12971284
1298-
In the multilabel case with binary indicator format:
1285+
In the multilabel case with binary label indicators:
12991286
1300-
>>> accuracy_score(np.array([[0.0, 1.0], [1.0, 1.0]]), np.ones((2, 2)))
1287+
>>> accuracy_score(np.array([[0, 1], [1, 1]]), np.ones((2, 2)))
13011288
0.5
1302-
1303-
and with a list of labels format:
1304-
1305-
>>> accuracy_score([(1, ), (3, )], [(1, 2), tuple()])
1306-
0.0
1307-
13081289
"""
13091290

13101291
# Compute accuracy for each possible representation
@@ -1343,10 +1324,10 @@ def f1_score(y_true, y_pred, labels=None, pos_label=1, average='weighted',
13431324
13441325
Parameters
13451326
----------
1346-
y_true : array-like or list of labels or label indicator matrix
1327+
y_true : array-like or label indicator matrix
13471328
Ground truth (correct) target values.
13481329
1349-
y_pred : array-like or list of labels or label indicator matrix
1330+
y_pred : array-like or label indicator matrix
13501331
Estimated targets as returned by a classifier.
13511332
13521333
labels : array
@@ -1426,10 +1407,10 @@ def fbeta_score(y_true, y_pred, beta, labels=None, pos_label=1,
14261407
14271408
Parameters
14281409
----------
1429-
y_true : array-like or list of labels or label indicator matrix
1410+
y_true : array-like or label indicator matrix
14301411
Ground truth (correct) target values.
14311412
1432-
y_pred : array-like or list of labels or label indicator matrix
1413+
y_pred : array-like or label indicator matrix
14331414
Estimated targets as returned by a classifier.
14341415
14351416
beta: float
@@ -1585,10 +1566,10 @@ def precision_recall_fscore_support(y_true, y_pred, beta=1.0, labels=None,
15851566
15861567
Parameters
15871568
----------
1588-
y_true : array-like or list of labels or label indicator matrix
1569+
y_true : array-like or label indicator matrix
15891570
Ground truth (correct) target values.
15901571
1591-
y_pred : array-like or list of labels or label indicator matrix
1572+
y_pred : array-like or label indicator matrix
15921573
Estimated targets as returned by a classifier.
15931574
15941575
beta : float, 1.0 by default
@@ -1830,10 +1811,10 @@ def precision_score(y_true, y_pred, labels=None, pos_label=1,
18301811
18311812
Parameters
18321813
----------
1833-
y_true : array-like or list of labels or label indicator matrix
1814+
y_true : array-like or label indicator matrix
18341815
Ground truth (correct) target values.
18351816
1836-
y_pred : array-like or list of labels or label indicator matrix
1817+
y_pred : array-like or label indicator matrix
18371818
Estimated targets as returned by a classifier.
18381819
18391820
labels : array
@@ -1912,10 +1893,10 @@ def recall_score(y_true, y_pred, labels=None, pos_label=1, average='weighted',
19121893
19131894
Parameters
19141895
----------
1915-
y_true : array-like or list of labels or label indicator matrix
1896+
y_true : array-like or label indicator matrix
19161897
Ground truth (correct) target values.
19171898
1918-
y_pred : array-like or list of labels or label indicator matrix
1899+
y_pred : array-like or label indicator matrix
19191900
Estimated targets as returned by a classifier.
19201901
19211902
labels : array
@@ -1987,10 +1968,10 @@ def classification_report(y_true, y_pred, labels=None, target_names=None,
19871968
19881969
Parameters
19891970
----------
1990-
y_true : array-like or list of labels or label indicator matrix
1971+
y_true : array-like or label indicator matrix
19911972
Ground truth (correct) target values.
19921973
1993-
y_pred : array-like or list of labels or label indicator matrix
1974+
y_pred : array-like or label indicator matrix
19941975
Estimated targets as returned by a classifier.
19951976
19961977
labels : array, shape = [n_labels]
@@ -2081,10 +2062,10 @@ def hamming_loss(y_true, y_pred, classes=None):
20812062
20822063
Parameters
20832064
----------
2084-
y_true : array-like or list of labels or label indicator matrix
2065+
y_true : array-like or label indicator matrix
20852066
Ground truth (correct) labels.
20862067
2087-
y_pred : array-like or list of labels or label indicator matrix
2068+
y_pred : array-like or label indicator matrix
20882069
Predicted labels, as returned by a classifier.
20892070
20902071
classes : array, shape = [n_labels], optional
@@ -2132,16 +2113,10 @@ def hamming_loss(y_true, y_pred, classes=None):
21322113
>>> hamming_loss(y_true, y_pred)
21332114
0.25
21342115
2135-
In the multilabel case with binary indicator format:
2116+
In the multilabel case with binary label indicators:
21362117
2137-
>>> hamming_loss(np.array([[0.0, 1.0], [1.0, 1.0]]), np.zeros((2, 2)))
2118+
>>> hamming_loss(np.array([[0, 1], [1, 1]]), np.zeros((2, 2)))
21382119
0.75
2139-
2140-
and with a list of labels format:
2141-
2142-
>>> hamming_loss([(1, 2), (3, )], [(1, 2), tuple()]) # doctest: +ELLIPSIS
2143-
0.166...
2144-
21452120
"""
21462121
y_type, y_true, y_pred = _check_clf_targets(y_true, y_pred)
21472122

‎sklearn/multiclass.py

Lines changed: 1 addition & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -143,9 +143,7 @@ class OneVsRestClassifier(BaseEstimator, ClassifierMixin, MetaEstimatorMixin):
143143
144144
This strategy can also be used for multilabel learning, where a classifier
145145
is used to predict multiple labels for instance, by fitting on a 2-d matrix
146-
in which cell [i, j] is 1 sample i has label j and 0 otherwise.
147-
For multilabel learning, the number of classes must be at
148-
least three, since otherwise OvR reduces to binary classification.
146+
in which cell [i, j] is 1 if sample i has label j and 0 otherwise.
149147
150148
In the multilabel learning literature, OvR is also known as the binary
151149
relevance method.

‎sklearn/preprocessing/label.py

Lines changed: 4 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -249,8 +249,7 @@ def fit(self, y):
249249
----------
250250
y : numpy array of shape (n_samples,) or (n_samples, n_classes)
251251
Target values. The 2-d matrix should only contain 0 and 1,
252-
represents multilabel classification, and is returned unchanged
253-
by LabelBinarizer.
252+
represents multilabel classification.
254253
255254
Returns
256255
-------
@@ -275,8 +274,7 @@ def transform(self, y):
275274
----------
276275
y : numpy array of shape (n_samples,) or (n_samples, n_classes)
277276
Target values. The 2-d matrix should only contain 0 and 1,
278-
represents multilabel classification, and is returned unchanged
279-
by LabelBinarizer.
277+
represents multilabel classification.
280278
281279
Returns
282280
-------
@@ -318,8 +316,7 @@ def inverse_transform(self, Y, threshold=None):
318316
-------
319317
y : numpy array of shape (n_samples,) or (n_samples, n_classes)
320318
Target values. The 2-d matrix should only contain 0 and 1,
321-
represents multilabel classification, and is returned unchanged
322-
by LabelBinarizer.
319+
represents multilabel classification.
323320
324321
Notes
325322
-----
@@ -569,8 +566,8 @@ def _transform(self, y, class_mapping):
569566
for labels in y:
570567
indices.extend(set(class_mapping[label] for label in labels))
571568
indptr.append(len(indices))
572-
# virtual array of len(indices) 1s:
573569
data = np.ones(len(indices), dtype=int)
570+
574571
return sp.csr_matrix((data, indices, indptr),
575572
shape=(len(indptr) - 1, len(class_mapping)))
576573

0 commit comments

Comments
 (0)