def test_base_estimator():
# Test different base estimators.
from sklearn.ensemble import RandomForestClassifier
# XXX doesn't work with y_class because RF doesn't support classes_
# Shouldn't AdaBoost run a LabelBinarizer?
clf = AdaBoostClassifier(RandomForestClassifier())
clf.fit(X, y_regr)
clf = AdaBoostClassifier(SVC(), algorithm="SAMME")
clf.fit(X, y_class)
from sklearn.ensemble import RandomForestRegressor
clf = AdaBoostRegressor(RandomForestRegressor(), random_state=0)
clf.fit(X, y_regr)
clf = AdaBoostRegressor(SVR(), random_state=0)
clf.fit(X, y_regr)
# Check that an empty discrete ensemble fails in fit, not predict.
X_fail = [[1, 1], [1, 1], [1, 1], [1, 1]]
y_fail = ["foo", "bar", 1, 2]
clf = AdaBoostClassifier(SVC(), algorithm="SAMME")
assert_raises_regexp(ValueError, "worse than random",
clf.fit, X_fail, y_fail)
def test_ovr_partial_fit_exceptions():
ovr = OneVsRestClassifier(MultinomialNB())
X = np.abs(np.random.randn(14, 2))
y = [1, 1, 1, 1, 2, 3, 3, 0, 0, 2, 3, 1, 2, 3]
ovr.partial_fit(X[:7], y[:7], np.unique(y))
# A new class value which was not in the first call of partial_fit
# It should raise ValueError
y1 = [5] + y[7:-1]
assert_raises_regexp(ValueError, r"Mini-batch contains \[.+\] while "
r"classes must be subset of \[.+\]",
ovr.partial_fit, X=X[7:], y=y1)
def test_ransac_resid_thresh_no_inliers():
# When residual_threshold=0.0 there are no inliers and a
# ValueError with a message should be raised
base_estimator = LinearRegression()
ransac_estimator = RANSACRegressor(base_estimator, min_samples=2,
residual_threshold=0.0, random_state=0,
max_trials=5)
msg = ("RANSAC could not find a valid consensus set")
assert_raises_regexp(ValueError, msg, ransac_estimator.fit, X, y)
assert ransac_estimator.n_skips_no_inliers_ == 5
assert ransac_estimator.n_skips_invalid_data_ == 0
assert ransac_estimator.n_skips_invalid_model_ == 0
def test_ransac_no_valid_model():
def is_model_valid(estimator, X, y):
return False
base_estimator = LinearRegression()
ransac_estimator = RANSACRegressor(base_estimator,
is_model_valid=is_model_valid,
max_trials=5)
msg = ("RANSAC could not find a valid consensus set")
assert_raises_regexp(ValueError, msg, ransac_estimator.fit, X, y)
assert ransac_estimator.n_skips_no_inliers_ == 0
assert ransac_estimator.n_skips_invalid_data_ == 0
assert ransac_estimator.n_skips_invalid_model_ == 5
def test_ovo_partial_fit_predict():
temp = datasets.load_iris()
X, y = temp.data, temp.target
ovo1 = OneVsOneClassifier(MultinomialNB())
ovo1.partial_fit(X[:100], y[:100], np.unique(y))
ovo1.partial_fit(X[100:], y[100:])
pred1 = ovo1.predict(X)
ovo2 = OneVsOneClassifier(MultinomialNB())
ovo2.fit(X, y)
pred2 = ovo2.predict(X)
assert len(ovo1.estimators_) == n_classes * (n_classes - 1) / 2
assert np.mean(y == pred1) > 0.65
assert_almost_equal(pred1, pred2)
# Test when mini-batches have binary target classes
ovo1 = OneVsOneClassifier(MultinomialNB())
ovo1.partial_fit(X[:60], y[:60], np.unique(y))
ovo1.partial_fit(X[60:], y[60:])
pred1 = ovo1.predict(X)
ovo2 = OneVsOneClassifier(MultinomialNB())
pred2 = ovo2.fit(X, y).predict(X)
assert_almost_equal(pred1, pred2)
assert len(ovo1.estimators_) == len(np.unique(y))
assert np.mean(y == pred1) > 0.65
ovo = OneVsOneClassifier(MultinomialNB())
X = np.random.rand(14, 2)
y = [1, 1, 2, 3, 3, 0, 0, 4, 4, 4, 4, 4, 2, 2]
ovo.partial_fit(X[:7], y[:7], [0, 1, 2, 3, 4])
ovo.partial_fit(X[7:], y[7:])
pred = ovo.predict(X)
ovo2 = OneVsOneClassifier(MultinomialNB())
pred2 = ovo2.fit(X, y).predict(X)
assert_almost_equal(pred, pred2)
# raises error when mini-batch does not have classes from all_classes
ovo = OneVsOneClassifier(MultinomialNB())
error_y = [0, 1, 2, 3, 4, 5, 2]
message_re = escape("Mini-batch contains {0} while "
"it must be subset of {1}".format(
np.unique(error_y), np.unique(y)))
assert_raises_regexp(ValueError, message_re, ovo.partial_fit, X[:7],
error_y, np.unique(y))
# test partial_fit only exists if estimator has it:
ovr = OneVsOneClassifier(SVC())
assert not hasattr(ovr, "partial_fit")
def test_ransac_exceed_max_skips():
def is_data_valid(X, y):
return False
base_estimator = LinearRegression()
ransac_estimator = RANSACRegressor(base_estimator,
is_data_valid=is_data_valid,
max_trials=5,
max_skips=3)
msg = ("RANSAC skipped more iterations than `max_skips`")
assert_raises_regexp(ValueError, msg, ransac_estimator.fit, X, y)
assert ransac_estimator.n_skips_no_inliers_ == 0
assert ransac_estimator.n_skips_invalid_data_ == 4
assert ransac_estimator.n_skips_invalid_model_ == 0
def test_check_1d_vectors():
# test check_1d_vectors function
rng = np.random.RandomState(0)
dim = rng.randint(50, 100)
# check (dim, )
a = _check_1d_vector(np.ones((dim, )))
assert_equal(1, len(a.shape))
assert_equal(dim, a.shape[0])
# check (dim, 1)
b = _check_1d_vector(np.ones((dim, 1)))
assert_equal(1, len(b.shape))
assert_equal(dim, b.shape[0])
# check (dim, 2)
c = np.ones((dim, 2))
assert_raises_regexp(ValueError, r"^Vector is not 1-d array:",
_check_1d_vector, c)
def test_group_kfold():
rng = np.random.RandomState(0)
# Parameters of the test
n_groups = 15
n_samples = 1000
n_splits = 5
X = y = np.ones(n_samples)
# Construct the test data
tolerance = 0.05 * n_samples # 5 percent error allowed
groups = rng.randint(0, n_groups, n_samples)
ideal_n_groups_per_fold = n_samples // n_splits
len(np.unique(groups))
# Get the test fold indices from the test set indices of each fold
folds = np.zeros(n_samples)
lkf = StratifiedGroupKFold(n_splits=n_splits)
for i, (_, test) in enumerate(lkf.split(X, y, groups)):
folds[test] = i
# Check that folds have approximately the same size
assert len(folds) == len(groups)
for i in np.unique(folds):
assert (tolerance >= abs(sum(folds == i) - ideal_n_groups_per_fold))
# Check that each group appears only in 1 fold
for group in np.unique(groups):
assert len(np.unique(folds[groups == group])) == 1
# Check that no group is on both sides of the split
groups = np.asarray(groups, dtype=object)
for train, test in lkf.split(X, y, groups):
assert len(np.intersect1d(groups[train], groups[test])) == 0
# Construct the test data
groups = np.array([
'Albert', 'Jean', 'Bertrand', 'Michel', 'Jean', 'Francis', 'Robert',
'Michel', 'Rachel', 'Lois', 'Michelle', 'Bernard', 'Marion', 'Laura',
'Jean', 'Rachel', 'Franck', 'John', 'Gael', 'Anna', 'Alix', 'Robert',
'Marion', 'David', 'Tony', 'Abel', 'Becky', 'Madmood', 'Cary', 'Mary',
'Alexandre', 'David', 'Francis', 'Barack', 'Abdoul', 'Rasha', 'Xi',
'Silvia'
])
n_groups = len(np.unique(groups))
n_samples = len(groups)
n_splits = 5
tolerance = 0.05 * n_samples # 5 percent error allowed
ideal_n_groups_per_fold = n_samples // n_splits
X = y = np.ones(n_samples)
# Get the test fold indices from the test set indices of each fold
folds = np.zeros(n_samples)
for i, (_, test) in enumerate(lkf.split(X, y, groups)):
folds[test] = i
# Check that folds have approximately the same size
assert len(folds) == len(groups)
for i in np.unique(folds):
assert (tolerance >= abs(sum(folds == i) - ideal_n_groups_per_fold))
# Check that each group appears only in 1 fold
with warnings.catch_warnings():
warnings.simplefilter("ignore", FutureWarning)
for group in np.unique(groups):
assert len(np.unique(folds[groups == group])) == 1
# Check that no group is on both sides of the split
groups = np.asarray(groups, dtype=object)
for train, test in lkf.split(X, y, groups):
assert len(np.intersect1d(groups[train], groups[test])) == 0
# groups can also be a list
cv_iter = list(lkf.split(X, y, groups.tolist()))
for (train1, test1), (train2, test2) in zip(lkf.split(X, y, groups),
cv_iter):
assert_array_equal(train1, train2)
assert_array_equal(test1, test2)
# Should fail if there are more folds than groups
groups = np.array([1, 1, 1, 2, 2])
X = y = np.ones(len(groups))
assert_raises_regexp(ValueError, "Cannot have number of splits.*greater",
next,
StratifiedGroupKFold(n_splits=3).split(X, y, groups))