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
|
# Author: Alexander Fabisch <afabisch@informatik.uni-bremen.de>
#
# License: BSD 3 clause
import sys
from sklearn.externals.six.moves import cStringIO as StringIO
import numpy as np
import warnings
from sklearn.base import BaseEstimator
from sklearn.utils.testing import assert_raises
from sklearn.utils.testing import assert_warns
from sklearn.utils.testing import assert_equal
from sklearn.utils.testing import assert_array_equal
from sklearn.utils.testing import assert_array_almost_equal
from sklearn.datasets import make_classification
with warnings.catch_warnings():
warnings.simplefilter('ignore')
from sklearn.learning_curve import learning_curve, validation_curve
from sklearn.cross_validation import KFold
from sklearn.linear_model import PassiveAggressiveClassifier
class MockImprovingEstimator(BaseEstimator):
"""Dummy classifier to test the learning curve"""
def __init__(self, n_max_train_sizes):
self.n_max_train_sizes = n_max_train_sizes
self.train_sizes = 0
self.X_subset = None
def fit(self, X_subset, y_subset=None):
self.X_subset = X_subset
self.train_sizes = X_subset.shape[0]
return self
def predict(self, X):
raise NotImplementedError
def score(self, X=None, Y=None):
# training score becomes worse (2 -> 1), test error better (0 -> 1)
if self._is_training_data(X):
return 2. - float(self.train_sizes) / self.n_max_train_sizes
else:
return float(self.train_sizes) / self.n_max_train_sizes
def _is_training_data(self, X):
return X is self.X_subset
class MockIncrementalImprovingEstimator(MockImprovingEstimator):
"""Dummy classifier that provides partial_fit"""
def __init__(self, n_max_train_sizes):
super(MockIncrementalImprovingEstimator,
self).__init__(n_max_train_sizes)
self.x = None
def _is_training_data(self, X):
return self.x in X
def partial_fit(self, X, y=None, **params):
self.train_sizes += X.shape[0]
self.x = X[0]
class MockEstimatorWithParameter(BaseEstimator):
"""Dummy classifier to test the validation curve"""
def __init__(self, param=0.5):
self.X_subset = None
self.param = param
def fit(self, X_subset, y_subset):
self.X_subset = X_subset
self.train_sizes = X_subset.shape[0]
return self
def predict(self, X):
raise NotImplementedError
def score(self, X=None, y=None):
return self.param if self._is_training_data(X) else 1 - self.param
def _is_training_data(self, X):
return X is self.X_subset
def test_learning_curve():
X, y = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockImprovingEstimator(20)
with warnings.catch_warnings(record=True) as w:
train_sizes, train_scores, test_scores = learning_curve(
estimator, X, y, cv=3, train_sizes=np.linspace(0.1, 1.0, 10))
if len(w) > 0:
raise RuntimeError("Unexpected warning: %r" % w[0].message)
assert_equal(train_scores.shape, (10, 3))
assert_equal(test_scores.shape, (10, 3))
assert_array_equal(train_sizes, np.linspace(2, 20, 10))
assert_array_almost_equal(train_scores.mean(axis=1),
np.linspace(1.9, 1.0, 10))
assert_array_almost_equal(test_scores.mean(axis=1),
np.linspace(0.1, 1.0, 10))
def test_learning_curve_unsupervised():
X, _ = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockImprovingEstimator(20)
train_sizes, train_scores, test_scores = learning_curve(
estimator, X, y=None, cv=3, train_sizes=np.linspace(0.1, 1.0, 10))
assert_array_equal(train_sizes, np.linspace(2, 20, 10))
assert_array_almost_equal(train_scores.mean(axis=1),
np.linspace(1.9, 1.0, 10))
assert_array_almost_equal(test_scores.mean(axis=1),
np.linspace(0.1, 1.0, 10))
def test_learning_curve_verbose():
X, y = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockImprovingEstimator(20)
old_stdout = sys.stdout
sys.stdout = StringIO()
try:
train_sizes, train_scores, test_scores = \
learning_curve(estimator, X, y, cv=3, verbose=1)
finally:
out = sys.stdout.getvalue()
sys.stdout.close()
sys.stdout = old_stdout
assert("[learning_curve]" in out)
def test_learning_curve_incremental_learning_not_possible():
X, y = make_classification(n_samples=2, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
# The mockup does not have partial_fit()
estimator = MockImprovingEstimator(1)
assert_raises(ValueError, learning_curve, estimator, X, y,
exploit_incremental_learning=True)
def test_learning_curve_incremental_learning():
X, y = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockIncrementalImprovingEstimator(20)
train_sizes, train_scores, test_scores = learning_curve(
estimator, X, y, cv=3, exploit_incremental_learning=True,
train_sizes=np.linspace(0.1, 1.0, 10))
assert_array_equal(train_sizes, np.linspace(2, 20, 10))
assert_array_almost_equal(train_scores.mean(axis=1),
np.linspace(1.9, 1.0, 10))
assert_array_almost_equal(test_scores.mean(axis=1),
np.linspace(0.1, 1.0, 10))
def test_learning_curve_incremental_learning_unsupervised():
X, _ = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockIncrementalImprovingEstimator(20)
train_sizes, train_scores, test_scores = learning_curve(
estimator, X, y=None, cv=3, exploit_incremental_learning=True,
train_sizes=np.linspace(0.1, 1.0, 10))
assert_array_equal(train_sizes, np.linspace(2, 20, 10))
assert_array_almost_equal(train_scores.mean(axis=1),
np.linspace(1.9, 1.0, 10))
assert_array_almost_equal(test_scores.mean(axis=1),
np.linspace(0.1, 1.0, 10))
def test_learning_curve_batch_and_incremental_learning_are_equal():
X, y = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
train_sizes = np.linspace(0.2, 1.0, 5)
estimator = PassiveAggressiveClassifier(n_iter=1, shuffle=False)
train_sizes_inc, train_scores_inc, test_scores_inc = \
learning_curve(
estimator, X, y, train_sizes=train_sizes,
cv=3, exploit_incremental_learning=True)
train_sizes_batch, train_scores_batch, test_scores_batch = \
learning_curve(
estimator, X, y, cv=3, train_sizes=train_sizes,
exploit_incremental_learning=False)
assert_array_equal(train_sizes_inc, train_sizes_batch)
assert_array_almost_equal(train_scores_inc.mean(axis=1),
train_scores_batch.mean(axis=1))
assert_array_almost_equal(test_scores_inc.mean(axis=1),
test_scores_batch.mean(axis=1))
def test_learning_curve_n_sample_range_out_of_bounds():
X, y = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockImprovingEstimator(20)
assert_raises(ValueError, learning_curve, estimator, X, y, cv=3,
train_sizes=[0, 1])
assert_raises(ValueError, learning_curve, estimator, X, y, cv=3,
train_sizes=[0.0, 1.0])
assert_raises(ValueError, learning_curve, estimator, X, y, cv=3,
train_sizes=[0.1, 1.1])
assert_raises(ValueError, learning_curve, estimator, X, y, cv=3,
train_sizes=[0, 20])
assert_raises(ValueError, learning_curve, estimator, X, y, cv=3,
train_sizes=[1, 21])
def test_learning_curve_remove_duplicate_sample_sizes():
X, y = make_classification(n_samples=3, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockImprovingEstimator(2)
train_sizes, _, _ = assert_warns(
RuntimeWarning, learning_curve, estimator, X, y, cv=3,
train_sizes=np.linspace(0.33, 1.0, 3))
assert_array_equal(train_sizes, [1, 2])
def test_learning_curve_with_boolean_indices():
X, y = make_classification(n_samples=30, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
estimator = MockImprovingEstimator(20)
cv = KFold(n=30, n_folds=3)
train_sizes, train_scores, test_scores = learning_curve(
estimator, X, y, cv=cv, train_sizes=np.linspace(0.1, 1.0, 10))
assert_array_equal(train_sizes, np.linspace(2, 20, 10))
assert_array_almost_equal(train_scores.mean(axis=1),
np.linspace(1.9, 1.0, 10))
assert_array_almost_equal(test_scores.mean(axis=1),
np.linspace(0.1, 1.0, 10))
def test_validation_curve():
X, y = make_classification(n_samples=2, n_features=1, n_informative=1,
n_redundant=0, n_classes=2,
n_clusters_per_class=1, random_state=0)
param_range = np.linspace(0, 1, 10)
with warnings.catch_warnings(record=True) as w:
train_scores, test_scores = validation_curve(
MockEstimatorWithParameter(), X, y, param_name="param",
param_range=param_range, cv=2
)
if len(w) > 0:
raise RuntimeError("Unexpected warning: %r" % w[0].message)
assert_array_almost_equal(train_scores.mean(axis=1), param_range)
assert_array_almost_equal(test_scores.mean(axis=1), 1 - param_range)
|
