def test_group_shuffle_split():
for groups_i in test_groups:
X = y = np.ones(len(groups_i))
n_splits = 6
test_size = 1./3
slo = GroupShuffleSplit(n_splits, test_size=test_size, random_state=0)
# Make sure the repr works
repr(slo)
# Test that the length is correct
assert_equal(slo.get_n_splits(X, y, groups=groups_i), n_splits)
l_unique = np.unique(groups_i)
l = np.asarray(groups_i)
for train, test in slo.split(X, y, groups=groups_i):
# First test: no train group is in the test set and vice versa
l_train_unique = np.unique(l[train])
l_test_unique = np.unique(l[test])
assert_false(np.any(np.in1d(l[train], l_test_unique)))
assert_false(np.any(np.in1d(l[test], l_train_unique)))
# Second test: train and test add up to all the data
assert_equal(l[train].size + l[test].size, l.size)
# Third test: train and test are disjoint
assert_array_equal(np.intersect1d(train, test), [])
# Fourth test:
# unique train and test groups are correct, +- 1 for rounding error
assert_true(abs(len(l_test_unique) -
round(test_size * len(l_unique))) <= 1)
assert_true(abs(len(l_train_unique) -
round((1.0 - test_size) * len(l_unique))) <= 1)
def test_predict_proba_disabled():
"""Test predict_proba when disabled on estimator."""
X = np.arange(20).reshape(5, -1)
y = [0, 0, 1, 1, 1]
clf = SVC(probability=False)
gs = GridSearchCV(clf, {}, cv=2).fit(X, y)
assert_false(hasattr(gs, "predict_proba"))
def fit(self, X_subset, y_subset):
assert_false(
hasattr(self, 'fit_called_'),
'fit is called the second time'
)
self.fit_called_ = True
return super(type(self), self).fit(X_subset, y_subset)
def test_lda_store_covariance():
# Test for slover 'lsqr' and 'eigen'
# 'store_covariance' has no effect on 'lsqr' and 'eigen' solvers
for solver in ('lsqr', 'eigen'):
clf = LinearDiscriminantAnalysis(solver=solver).fit(X6, y6)
assert_true(hasattr(clf, 'covariance_'))
# Test the actual attribute:
clf = LinearDiscriminantAnalysis(solver=solver,
store_covariance=True).fit(X6, y6)
assert_true(hasattr(clf, 'covariance_'))
assert_array_almost_equal(
clf.covariance_,
np.array([[0.422222, 0.088889], [0.088889, 0.533333]])
)
# Test for SVD slover, the default is to not set the covariances_ attribute
clf = LinearDiscriminantAnalysis(solver='svd').fit(X6, y6)
assert_false(hasattr(clf, 'covariance_'))
# Test the actual attribute:
clf = LinearDiscriminantAnalysis(solver=solver,
store_covariance=True).fit(X6, y6)
assert_true(hasattr(clf, 'covariance_'))
assert_array_almost_equal(
clf.covariance_,
np.array([[0.422222, 0.088889], [0.088889, 0.533333]])
)
def test_as_float_array():
# Test function for as_float_array
X = np.ones((3, 10), dtype=np.int32)
X = X + np.arange(10, dtype=np.int32)
# Checks that the return type is ok
X2 = as_float_array(X, copy=False)
np.testing.assert_equal(X2.dtype, np.float32)
# Another test
X = X.astype(np.int64)
X2 = as_float_array(X, copy=True)
# Checking that the array wasn't overwritten
assert_true(as_float_array(X, False) is not X)
# Checking that the new type is ok
np.testing.assert_equal(X2.dtype, np.float64)
# Here, X is of the right type, it shouldn't be modified
X = np.ones((3, 2), dtype=np.float32)
assert_true(as_float_array(X, copy=False) is X)
# Test that if X is fortran ordered it stays
X = np.asfortranarray(X)
assert_true(np.isfortran(as_float_array(X, copy=True)))
# Test the copy parameter with some matrices
matrices = [
np.matrix(np.arange(5)),
sp.csc_matrix(np.arange(5)).toarray(),
sparse_random_matrix(10, 10, density=0.10).toarray()
]
for M in matrices:
N = as_float_array(M, copy=True)
N[0, 0] = np.nan
assert_false(np.isnan(M).any())
def test_unsorted_indices():
# test that the result with sorted and unsorted indices in csr is the same
# we use a subset of digits as iris, blobs or make_classification didn't
# show the problem
digits = load_digits()
X, y = digits.data[:50], digits.target[:50]
X_test = sparse.csr_matrix(digits.data[50:100])
X_sparse = sparse.csr_matrix(X)
coef_dense = svm.SVC(kernel='linear', probability=True,
random_state=0).fit(X, y).coef_
sparse_svc = svm.SVC(kernel='linear', probability=True,
random_state=0).fit(X_sparse, y)
coef_sorted = sparse_svc.coef_
# make sure dense and sparse SVM give the same result
assert_array_almost_equal(coef_dense, coef_sorted.toarray())
X_sparse_unsorted = X_sparse[np.arange(X.shape[0])]
X_test_unsorted = X_test[np.arange(X_test.shape[0])]
# make sure we scramble the indices
assert_false(X_sparse_unsorted.has_sorted_indices)
assert_false(X_test_unsorted.has_sorted_indices)
unsorted_svc = svm.SVC(kernel='linear', probability=True,
random_state=0).fit(X_sparse_unsorted, y)
coef_unsorted = unsorted_svc.coef_
# make sure unsorted indices give same result
assert_array_almost_equal(coef_unsorted.toarray(), coef_sorted.toarray())
assert_array_almost_equal(sparse_svc.predict_proba(X_test_unsorted),
sparse_svc.predict_proba(X_test))
def test_scale_function_without_centering():
rng = np.random.RandomState(42)
X = rng.randn(4, 5)
X[:, 0] = 0.0 # first feature is always of zero
X_csr = sparse.csr_matrix(X)
X_scaled = scale(X, with_mean=False)
assert_false(np.any(np.isnan(X_scaled)))
X_csr_scaled = scale(X_csr, with_mean=False)
assert_false(np.any(np.isnan(X_csr_scaled.data)))
# test csc has same outcome
X_csc_scaled = scale(X_csr.tocsc(), with_mean=False)
assert_array_almost_equal(X_scaled, X_csc_scaled.toarray())
# raises value error on axis != 0
assert_raises(ValueError, scale, X_csr, with_mean=False, axis=1)
assert_array_almost_equal(X_scaled.mean(axis=0),
[0., -0.01, 2.24, -0.35, -0.78], 2)
assert_array_almost_equal(X_scaled.std(axis=0), [0., 1., 1., 1., 1.])
# Check that X has not been copied
assert_true(X_scaled is not X)
X_csr_scaled_mean, X_csr_scaled_std = mean_variance_axis0(X_csr_scaled)
assert_array_almost_equal(X_csr_scaled_mean, X_scaled.mean(axis=0))
assert_array_almost_equal(X_csr_scaled_std, X_scaled.std(axis=0))
def test_pickle_version_warning():
# check that warnings are raised when unpickling in a different version
# first, check no warning when in the same version:
iris = datasets.load_iris()
tree = DecisionTreeClassifier().fit(iris.data, iris.target)
tree_pickle = pickle.dumps(tree)
assert_true(b"version" in tree_pickle)
assert_no_warnings(pickle.loads, tree_pickle)
# check that warning is raised on different version
tree_pickle_other = tree_pickle.replace(sklearn.__version__.encode(),
b"something")
message = ("Trying to unpickle estimator DecisionTreeClassifier from "
"version {0} when using version {1}. This might lead to "
"breaking code or invalid results. "
"Use at your own risk.".format("something",
sklearn.__version__))
assert_warns_message(UserWarning, message, pickle.loads, tree_pickle_other)
# check that not including any version also works:
# TreeNoVersion has no getstate, like pre-0.18
tree = TreeNoVersion().fit(iris.data, iris.target)
tree_pickle_noversion = pickle.dumps(tree)
assert_false(b"version" in tree_pickle_noversion)
message = message.replace("something", "pre-0.18")
message = message.replace("DecisionTreeClassifier", "TreeNoVersion")
# check we got the warning about using pre-0.18 pickle
assert_warns_message(UserWarning, message, pickle.loads,
tree_pickle_noversion)
# check that no warning is raised for external estimators
TreeNoVersion.__module__ = "notsklearn"
assert_no_warnings(pickle.loads, tree_pickle_noversion)
def test_sgd_optimizer_trigger_stopping():
params = [np.zeros(shape) for shape in shapes]
lr = 2e-6
optimizer = SGDOptimizer(params, lr, lr_schedule='adaptive')
assert_false(optimizer.trigger_stopping('', False))
assert_equal(lr / 5, optimizer.learning_rate)
assert_true(optimizer.trigger_stopping('', False))
def test_ovo_partial_fit_predict():
X, y = shuffle(iris.data, iris.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_equal(len(ovo1.estimators_), n_classes * (n_classes - 1) / 2)
assert_greater(np.mean(y == pred1), 0.65)
assert_almost_equal(pred1, pred2)
# Test when mini-batches don't have all target classes
ovo1 = OneVsOneClassifier(MultinomialNB())
ovo1.partial_fit(iris.data[:60], iris.target[:60], np.unique(iris.target))
ovo1.partial_fit(iris.data[60:], iris.target[60:])
pred1 = ovo1.predict(iris.data)
ovo2 = OneVsOneClassifier(MultinomialNB())
pred2 = ovo2.fit(iris.data, iris.target).predict(iris.data)
assert_almost_equal(pred1, pred2)
assert_equal(len(ovo1.estimators_), len(np.unique(iris.target)))
assert_greater(np.mean(iris.target == pred1), 0.65)
# test partial_fit only exists if estimator has it:
ovr = OneVsOneClassifier(SVC())
assert_false(hasattr(ovr, "partial_fit"))
def test_error():
# Test that it gives proper exception on deficient input.
X, y = iris.data, iris.target
base = DecisionTreeClassifier()
# Test max_samples
assert_raises(ValueError,
BaggingClassifier(base, max_samples=-1).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_samples=0.0).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_samples=2.0).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_samples=1000).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_samples="foobar").fit, X, y)
# Test max_features
assert_raises(ValueError,
BaggingClassifier(base, max_features=-1).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_features=0.0).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_features=2.0).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_features=5).fit, X, y)
assert_raises(ValueError,
BaggingClassifier(base, max_features="foobar").fit, X, y)
# Test support of decision_function
assert_false(hasattr(BaggingClassifier(base).fit(X, y), 'decision_function'))
def test_feature_names():
cv = CountVectorizer(max_df=0.5)
# test for Value error on unfitted/empty vocabulary
assert_raises(ValueError, cv.get_feature_names)
assert_false(cv.fixed_vocabulary_)
# test for vocabulary learned from data
X = cv.fit_transform(ALL_FOOD_DOCS)
n_samples, n_features = X.shape
assert_equal(len(cv.vocabulary_), n_features)
feature_names = cv.get_feature_names()
assert_equal(len(feature_names), n_features)
assert_array_equal(['beer', 'burger', 'celeri', 'coke', 'pizza',
'salad', 'sparkling', 'tomato', 'water'],
feature_names)
for idx, name in enumerate(feature_names):
assert_equal(idx, cv.vocabulary_.get(name))
# test for custom vocabulary
vocab = ['beer', 'burger', 'celeri', 'coke', 'pizza',
'salad', 'sparkling', 'tomato', 'water']
cv = CountVectorizer(vocabulary=vocab)
feature_names = cv.get_feature_names()
assert_array_equal(['beer', 'burger', 'celeri', 'coke', 'pizza', 'salad',
'sparkling', 'tomato', 'water'], feature_names)
assert_true(cv.fixed_vocabulary_)
for idx, name in enumerate(feature_names):
assert_equal(idx, cv.vocabulary_.get(name))
def test_nmf_fit_nn_output():
# Test that the decomposition does not contain negative values
A = np.c_[5 * np.ones(5) - np.arange(1, 6), 5 * np.ones(5) + np.arange(1, 6)]
for init in (None, "nndsvd", "nndsvda", "nndsvdar"):
model = NMF(n_components=2, init=init, random_state=0)
transf = model.fit_transform(A)
assert_false((model.components_ < 0).any() or (transf < 0).any())
def test_set_params():
"""set_params should be able to set estimators"""
clf1 = LogisticRegression(random_state=123, C=1.0)
clf2 = RandomForestClassifier(random_state=123, max_depth=None)
clf3 = GaussianNB()
eclf1 = VotingClassifier([('lr', clf1), ('rf', clf2)], voting='soft',
weights=[1, 2])
assert_true('lr' in eclf1.named_estimators)
assert_true(eclf1.named_estimators.lr is eclf1.estimators[0][1])
assert_true(eclf1.named_estimators.lr is eclf1.named_estimators['lr'])
eclf1.fit(X, y)
assert_true('lr' in eclf1.named_estimators_)
assert_true(eclf1.named_estimators_.lr is eclf1.estimators_[0])
assert_true(eclf1.named_estimators_.lr is eclf1.named_estimators_['lr'])
eclf2 = VotingClassifier([('lr', clf1), ('nb', clf3)], voting='soft',
weights=[1, 2])
eclf2.set_params(nb=clf2).fit(X, y)
assert_false(hasattr(eclf2, 'nb'))
assert_array_equal(eclf1.predict(X), eclf2.predict(X))
assert_array_almost_equal(eclf1.predict_proba(X), eclf2.predict_proba(X))
assert_equal(eclf2.estimators[0][1].get_params(), clf1.get_params())
assert_equal(eclf2.estimators[1][1].get_params(), clf2.get_params())
eclf1.set_params(lr__C=10.0)
eclf2.set_params(nb__max_depth=5)
assert_true(eclf1.estimators[0][1].get_params()['C'] == 10.0)
assert_true(eclf2.estimators[1][1].get_params()['max_depth'] == 5)
assert_equal(eclf1.get_params()["lr__C"],
eclf1.get_params()["lr"].get_params()['C'])
def test_column_transformer_get_set_params():
ct = ColumnTransformer([('trans1', StandardScaler(), [0]),
('trans2', StandardScaler(), [1])])
exp = {'n_jobs': 1,
'remainder': 'drop',
'trans1': ct.transformers[0][1],
'trans1__copy': True,
'trans1__with_mean': True,
'trans1__with_std': True,
'trans2': ct.transformers[1][1],
'trans2__copy': True,
'trans2__with_mean': True,
'trans2__with_std': True,
'transformers': ct.transformers,
'transformer_weights': None}
assert_dict_equal(ct.get_params(), exp)
ct.set_params(trans1__with_mean=False)
assert_false(ct.get_params()['trans1__with_mean'])
ct.set_params(trans1='passthrough')
exp = {'n_jobs': 1,
'remainder': 'drop',
'trans1': 'passthrough',
'trans2': ct.transformers[1][1],
'trans2__copy': True,
'trans2__with_mean': True,
'trans2__with_std': True,
'transformers': ct.transformers,
'transformer_weights': None}
assert_dict_equal(ct.get_params(), exp)
def check_estimators_overwrite_params(name, Estimator, X, y):
with warnings.catch_warnings(record=True):
# catch deprecation warnings
estimator = Estimator()
if hasattr(estimator, 'batch_size'):
# FIXME
# for MiniBatchDictLearning
estimator.batch_size = 1
if name in ['GaussianRandomProjection',
'SparseRandomProjection']:
# Due to the jl lemma and very few samples, the number
# of components of the random matrix projection will be
# greater
# than the number of features.
# So we impose a smaller number (avoid "auto" mode)
estimator = Estimator(n_components=1)
elif name == "SelectKBest":
estimator = Estimator(k=1)
set_random_state(estimator)
params = estimator.get_params()
estimator.fit(X, y)
new_params = estimator.get_params()
for k, v in params.items():
assert_false(np.any(new_params[k] != v),
"Estimator %s changes its parameter %s"
" from %s to %s during fit."
% (name, k, v, new_params[k]))
def test_check_increasing_down():
x = [0, 1, 2, 3, 4, 5]
y = [0, -1.5, -2.77, -8.99, -8.99, -50]
# Check that we got increasing=False and no warnings
is_increasing = assert_no_warnings(check_increasing, x, y)
assert_false(is_increasing)
def test_initialize_nn_output():
# Test that initialization does not return negative values
rng = np.random.mtrand.RandomState(42)
data = np.abs(rng.randn(10, 10))
for init in ('random', 'nndsvd', 'nndsvda', 'nndsvdar'):
W, H = nmf._initialize_nmf(data, 10, init=init, random_state=0)
assert_false((W < 0).any() or (H < 0).any())
def test_optimizer_copy(acq_func):
# Checks that the base estimator, the objective and target values
# are copied correctly.
base_estimator = ExtraTreesRegressor(random_state=2)
opt = Optimizer([(-2.0, 2.0)], base_estimator, acq_func=acq_func,
n_initial_points=1, acq_optimizer="sampling")
# run three iterations so that we have some points and objective values
if "ps" in acq_func:
opt.run(bench1_with_time, n_iter=3)
else:
opt.run(bench1, n_iter=3)
opt_copy = opt.copy()
copied_estimator = opt_copy.base_estimator_
if "ps" in acq_func:
assert_true(isinstance(copied_estimator, MultiOutputRegressor))
# check that the base_estimator is not wrapped multiple times
is_multi = isinstance(copied_estimator.estimator,
MultiOutputRegressor)
assert_false(is_multi)
else:
assert_false(isinstance(copied_estimator, MultiOutputRegressor))
assert_array_equal(opt_copy.Xi, opt.Xi)
assert_array_equal(opt_copy.yi, opt.yi)
def test_pipeline_memory():
iris = load_iris()
X = iris.data
y = iris.target
cachedir = mkdtemp()
try:
memory = Memory(cachedir=cachedir, verbose=10)
# Test with Transformer + SVC
clf = SVC(gamma='scale', probability=True, random_state=0)
transf = DummyTransf()
pipe = Pipeline([('transf', clone(transf)), ('svc', clf)])
cached_pipe = Pipeline([('transf', transf), ('svc', clf)],
memory=memory)
# Memoize the transformer at the first fit
cached_pipe.fit(X, y)
pipe.fit(X, y)
# Get the time stamp of the transformer in the cached pipeline
ts = cached_pipe.named_steps['transf'].timestamp_
# Check that cached_pipe and pipe yield identical results
assert_array_equal(pipe.predict(X), cached_pipe.predict(X))
assert_array_equal(pipe.predict_proba(X), cached_pipe.predict_proba(X))
assert_array_equal(pipe.predict_log_proba(X),
cached_pipe.predict_log_proba(X))
assert_array_equal(pipe.score(X, y), cached_pipe.score(X, y))
assert_array_equal(pipe.named_steps['transf'].means_,
cached_pipe.named_steps['transf'].means_)
assert_false(hasattr(transf, 'means_'))
# Check that we are reading the cache while fitting
# a second time
cached_pipe.fit(X, y)
# Check that cached_pipe and pipe yield identical results
assert_array_equal(pipe.predict(X), cached_pipe.predict(X))
assert_array_equal(pipe.predict_proba(X), cached_pipe.predict_proba(X))
assert_array_equal(pipe.predict_log_proba(X),
cached_pipe.predict_log_proba(X))
assert_array_equal(pipe.score(X, y), cached_pipe.score(X, y))
assert_array_equal(pipe.named_steps['transf'].means_,
cached_pipe.named_steps['transf'].means_)
assert_equal(ts, cached_pipe.named_steps['transf'].timestamp_)
# Create a new pipeline with cloned estimators
# Check that even changing the name step does not affect the cache hit
clf_2 = SVC(gamma='scale', probability=True, random_state=0)
transf_2 = DummyTransf()
cached_pipe_2 = Pipeline([('transf_2', transf_2), ('svc', clf_2)],
memory=memory)
cached_pipe_2.fit(X, y)
# Check that cached_pipe and pipe yield identical results
assert_array_equal(pipe.predict(X), cached_pipe_2.predict(X))
assert_array_equal(pipe.predict_proba(X),
cached_pipe_2.predict_proba(X))
assert_array_equal(pipe.predict_log_proba(X),
cached_pipe_2.predict_log_proba(X))
assert_array_equal(pipe.score(X, y), cached_pipe_2.score(X, y))
assert_array_equal(pipe.named_steps['transf'].means_,
cached_pipe_2.named_steps['transf_2'].means_)
assert_equal(ts, cached_pipe_2.named_steps['transf_2'].timestamp_)
finally:
shutil.rmtree(cachedir)
def test_check_is_permutation():
p = np.arange(100)
assert_true(_check_is_permutation(p, 100))
assert_false(_check_is_permutation(np.delete(p, 23), 100))
p[0] = 23
assert_false(_check_is_permutation(p, 100))
def test_check_is_partition():
p = np.arange(100)
assert_true(cval._check_is_partition(p, 100))
assert_false(cval._check_is_partition(np.delete(p, 23), 100))
p[0] = 23
assert_false(cval._check_is_partition(p, 100))
def test_randomized_svd_sign_flip_with_transpose():
# Check if the randomized_svd sign flipping is always done based on u
# irrespective of transpose.
# See https://github.com/scikit-learn/scikit-learn/issues/5608
# for more details.
def max_loading_is_positive(u, v):
"""
returns bool tuple indicating if the values maximising np.abs
are positive across all rows for u and across all columns for v.
"""
u_based = (np.abs(u).max(axis=0) == u.max(axis=0)).all()
v_based = (np.abs(v).max(axis=1) == v.max(axis=1)).all()
return u_based, v_based
mat = np.arange(10 * 8).reshape(10, -1)
# Without transpose
u_flipped, _, v_flipped = randomized_svd(mat, 3, flip_sign=True)
u_based, v_based = max_loading_is_positive(u_flipped, v_flipped)
assert_true(u_based)
assert_false(v_based)
# With transpose
u_flipped_with_transpose, _, v_flipped_with_transpose = randomized_svd(
mat, 3, flip_sign=True, transpose=True)
u_based, v_based = max_loading_is_positive(
u_flipped_with_transpose, v_flipped_with_transpose)
assert_true(u_based)
assert_false(v_based)
def test_mutual_info_options():
X = np.array([[0, 0, 0],
[1, 1, 0],
[2, 0, 1],
[2, 0, 1],
[2, 0, 1]], dtype=float)
y = np.array([0, 1, 2, 2, 1], dtype=float)
X_csr = csr_matrix(X)
for mutual_info in (mutual_info_regression, mutual_info_classif):
assert_raises(ValueError, mutual_info_regression, X_csr, y,
discrete_features=False)
mi_1 = mutual_info(X, y, discrete_features='auto', random_state=0)
mi_2 = mutual_info(X, y, discrete_features=False, random_state=0)
mi_3 = mutual_info(X_csr, y, discrete_features='auto',
random_state=0)
mi_4 = mutual_info(X_csr, y, discrete_features=True,
random_state=0)
assert_array_equal(mi_1, mi_2)
assert_array_equal(mi_3, mi_4)
assert_false(np.allclose(mi_1, mi_3))
def test_ovr_partial_fit():
# Test if partial_fit is working as intended
X, y = shuffle(iris.data, iris.target, random_state=0)
ovr = OneVsRestClassifier(MultinomialNB())
ovr.partial_fit(X[:100], y[:100], np.unique(y))
ovr.partial_fit(X[100:], y[100:])
pred = ovr.predict(X)
ovr2 = OneVsRestClassifier(MultinomialNB())
pred2 = ovr2.fit(X, y).predict(X)
assert_almost_equal(pred, pred2)
assert_equal(len(ovr.estimators_), len(np.unique(y)))
assert_greater(np.mean(y == pred), 0.65)
# Test when mini batches doesn't have all classes
# with SGDClassifier
X = np.abs(np.random.randn(14, 2))
y = [1, 1, 1, 1, 2, 3, 3, 0, 0, 2, 3, 1, 2, 3]
ovr = OneVsRestClassifier(SGDClassifier(max_iter=1, tol=None,
shuffle=False, random_state=0))
ovr.partial_fit(X[:7], y[:7], np.unique(y))
ovr.partial_fit(X[7:], y[7:])
pred = ovr.predict(X)
ovr1 = OneVsRestClassifier(SGDClassifier(max_iter=1, tol=None,
shuffle=False, random_state=0))
pred1 = ovr1.fit(X, y).predict(X)
assert_equal(np.mean(pred == y), np.mean(pred1 == y))
# test partial_fit only exists if estimator has it:
ovr = OneVsRestClassifier(SVC())
assert_false(hasattr(ovr, "partial_fit"))
def test_check_increasing_down_extreme():
x = [0, 1, 2, 3, 4, 5]
y = [0, -1, -2, -3, -4, -5]
# Check that we got increasing=False and no warnings
is_increasing = assert_no_warnings(check_increasing, x, y)
assert_false(is_increasing)
def check_estimators_overwrite_params(name, Estimator):
X, y = make_blobs(random_state=0, n_samples=9)
y = multioutput_estimator_convert_y_2d(name, y)
# some want non-negative input
X -= X.min()
with warnings.catch_warnings(record=True):
# catch deprecation warnings
estimator = Estimator()
if name == 'MiniBatchDictLearning' or name == 'MiniBatchSparsePCA':
# FIXME
# for MiniBatchDictLearning and MiniBatchSparsePCA
estimator.batch_size = 1
set_fast_parameters(estimator)
set_random_state(estimator)
params = estimator.get_params()
estimator.fit(X, y)
new_params = estimator.get_params()
for k, v in params.items():
assert_false(np.any(new_params[k] != v),
"Estimator %s changes its parameter %s"
" from %s to %s during fit."
% (name, k, v, new_params[k]))
def test_pipeline_init(self):
# Test the various init parameters of the pipeline.
assert_raises(TypeError, SparkPipeline)
# Check that we can't instantiate pipelines with objects without fit
# method
pipe = assert_raises(TypeError, SparkPipeline, [('svc', IncorrectT)])
# Smoke test with only an estimator
clf = T()
pipe = SparkPipeline([('svc', clf)])
assert_equal(pipe.get_params(deep=True),
dict(svc__a=None, svc__b=None, svc=clf,
**pipe.get_params(deep=False)
))
# Check that params are set
pipe.set_params(svc__a=0.1)
assert_equal(clf.a, 0.1)
assert_equal(clf.b, None)
# Smoke test the repr:
repr(pipe)
# Test with two objects
vect = SparkCountVectorizer()
filter = SparkVarianceThreshold()
pipe = SparkPipeline([('vect', vect), ('filter', filter)])
# Check that we can't use the same stage name twice
assert_raises(ValueError, SparkPipeline,
[('vect', vect), ('vect', vect)])
# Check that params are set
pipe.set_params(vect__min_df=0.1)
assert_equal(vect.min_df, 0.1)
# Smoke test the repr:
repr(pipe)
# Check that params are not set when naming them wrong
assert_raises(ValueError, pipe.set_params, filter__min_df=0.1)
# Test clone
pipe2 = clone(pipe)
assert_false(pipe.named_steps['vect'] is pipe2.named_steps['vect'])
# Check that apart from estimators, the parameters are the same
params = pipe.get_params(deep=True)
params2 = pipe2.get_params(deep=True)
for x in pipe.get_params(deep=False):
params.pop(x)
for x in pipe2.get_params(deep=False):
params2.pop(x)
# Remove estimators that where copied
params.pop('vect')
params.pop('filter')
params2.pop('vect')
params2.pop('filter')
assert_equal(params, params2)
def test_transform_nan():
# Test that SparsePCA won't return NaN when there is 0 feature in all
# samples.
rng = np.random.RandomState(0)
Y, _, _ = generate_toy_data(3, 10, (8, 8), random_state=rng) # wide array
Y[:, 0] = 0
estimator = SparsePCA(n_components=8)
assert_false(np.any(np.isnan(estimator.fit_transform(Y))))
def test_imputation_copy():
# Test imputation with copy
X_orig = sparse_random_matrix(5, 5, density=0.75, random_state=0)
# copy=True, dense => copy
X = X_orig.copy().toarray()
imputer = Imputer(missing_values=0, strategy="mean", copy=True)
Xt = imputer.fit(X).transform(X)
Xt[0, 0] = -1
assert_false(np.all(X == Xt))
# copy=True, sparse csr => copy
X = X_orig.copy()
imputer = Imputer(missing_values=X.data[0], strategy="mean", copy=True)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_false(np.all(X.data == Xt.data))
# copy=False, dense => no copy
X = X_orig.copy().toarray()
imputer = Imputer(missing_values=0, strategy="mean", copy=False)
Xt = imputer.fit(X).transform(X)
Xt[0, 0] = -1
assert_true(np.all(X == Xt))
# copy=False, sparse csr, axis=1 => no copy
X = X_orig.copy()
imputer = Imputer(missing_values=X.data[0], strategy="mean", copy=False, axis=1)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_true(np.all(X.data == Xt.data))
# copy=False, sparse csc, axis=0 => no copy
X = X_orig.copy().tocsc()
imputer = Imputer(missing_values=X.data[0], strategy="mean", copy=False, axis=0)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_true(np.all(X.data == Xt.data))
# copy=False, sparse csr, axis=0 => copy
X = X_orig.copy()
imputer = Imputer(missing_values=X.data[0], strategy="mean", copy=False, axis=0)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_false(np.all(X.data == Xt.data))
# copy=False, sparse csc, axis=1 => copy
X = X_orig.copy().tocsc()
imputer = Imputer(missing_values=X.data[0], strategy="mean", copy=False, axis=1)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_false(np.all(X.data == Xt.data))
# copy=False, sparse csr, axis=1, missing_values=0 => copy
X = X_orig.copy()
imputer = Imputer(missing_values=0, strategy="mean", copy=False, axis=1)
Xt = imputer.fit(X).transform(X)
assert_false(sparse.issparse(Xt))
def check_oob_score_raise_error(name):
ForestEstimator = FOREST_ESTIMATORS[name]
if name in FOREST_TRANSFORMERS:
for oob_score in [True, False]:
assert_raises(TypeError, ForestEstimator, oob_score=oob_score)
assert_raises(NotImplementedError,
ForestEstimator()._set_oob_score, X, y)
else:
# Unfitted / no bootstrap / no oob_score
for oob_score, bootstrap in [(True, False), (False, True),
(False, False)]:
est = ForestEstimator(oob_score=oob_score,
bootstrap=bootstrap,
random_state=0)
assert_false(hasattr(est, "oob_score_"))
# No bootstrap
assert_raises(ValueError,
ForestEstimator(oob_score=True, bootstrap=False).fit, X,
y)
def test_plain_has_no_average_attr(self):
clf = self.factory(average=True, eta0=.01)
clf.fit(X, Y)
assert_true(hasattr(clf, 'average_coef_'))
assert_true(hasattr(clf, 'average_intercept_'))
assert_true(hasattr(clf, 'standard_intercept_'))
assert_true(hasattr(clf, 'standard_coef_'))
clf = self.factory()
clf.fit(X, Y)
assert_false(hasattr(clf, 'average_coef_'))
assert_false(hasattr(clf, 'average_intercept_'))
assert_false(hasattr(clf, 'standard_intercept_'))
assert_false(hasattr(clf, 'standard_coef_'))
def test_mutual_info_options():
X = np.array([[0, 0, 0], [1, 1, 0], [2, 0, 1], [2, 0, 1], [2, 0, 1]],
dtype=float)
y = np.array([0, 1, 2, 2, 1], dtype=float)
X_csr = csr_matrix(X)
for mutual_info in (mutual_info_regression, mutual_info_classif):
assert_raises(ValueError,
mutual_info_regression,
X_csr,
y,
discrete_features=False)
mi_1 = mutual_info(X, y, discrete_features='auto', random_state=0)
mi_2 = mutual_info(X, y, discrete_features=False, random_state=0)
mi_3 = mutual_info(X_csr, y, discrete_features='auto', random_state=0)
mi_4 = mutual_info(X_csr, y, discrete_features=True, random_state=0)
assert_array_equal(mi_1, mi_2)
assert_array_equal(mi_3, mi_4)
assert_false(np.allclose(mi_1, mi_3))
def test_vectorizer_pipeline_grid_selection():
# raw documents
data = JUNK_FOOD_DOCS + NOTJUNK_FOOD_DOCS
# label junk food as -1, the others as +1
target = [-1] * len(JUNK_FOOD_DOCS) + [1] * len(NOTJUNK_FOOD_DOCS)
# split the dataset for model development and final evaluation
train_data, test_data, target_train, target_test = train_test_split(
data, target, test_size=.1, random_state=0)
pipeline = Pipeline([('vect', TfidfVectorizer()),
('svc', LinearSVC())])
parameters = {
'vect__ngram_range': [(1, 1), (1, 2)],
'vect__norm': ('l1', 'l2'),
'svc__loss': ('hinge', 'squared_hinge'),
}
# find the best parameters for both the feature extraction and the
# classifier
grid_search = GridSearchCV(pipeline, parameters, n_jobs=1)
# Check that the best model found by grid search is 100% correct on the
# held out evaluation set.
pred = grid_search.fit(train_data, target_train).predict(test_data)
assert_array_equal(pred, target_test)
# on this toy dataset bigram representation which is used in the last of
# the grid_search is considered the best estimator since they all converge
# to 100% accuracy models
assert_equal(grid_search.best_score_, 1.0)
best_vectorizer = grid_search.best_estimator_.named_steps['vect']
assert_equal(best_vectorizer.ngram_range, (1, 1))
assert_equal(best_vectorizer.norm, 'l2')
assert_false(best_vectorizer.fixed_vocabulary_)
def test_cv_iterable_wrapper():
y_multiclass = np.array([0, 1, 0, 1, 2, 1, 2, 0, 2])
with warnings.catch_warnings(record=True):
from sklearn.cross_validation import StratifiedKFold as OldSKF
cv = OldSKF(y_multiclass, n_folds=3)
wrapped_old_skf = _CVIterableWrapper(cv)
# Check if split works correctly
np.testing.assert_equal(list(cv), list(wrapped_old_skf.split()))
# Check if get_n_splits works correctly
assert_equal(len(cv), wrapped_old_skf.get_n_splits())
kf_iter = KFold(n_splits=5).split(X, y)
kf_iter_wrapped = check_cv(kf_iter)
# Since the wrapped iterable is enlisted and stored,
# split can be called any number of times to produce
# consistent results.
np.testing.assert_equal(list(kf_iter_wrapped.split(X, y)),
list(kf_iter_wrapped.split(X, y)))
# If the splits are randomized, successive calls to split yields different
# results
kf_randomized_iter = KFold(n_splits=5, shuffle=True).split(X, y)
kf_randomized_iter_wrapped = check_cv(kf_randomized_iter)
np.testing.assert_equal(list(kf_randomized_iter_wrapped.split(X, y)),
list(kf_randomized_iter_wrapped.split(X, y)))
try:
np.testing.assert_equal(list(kf_iter_wrapped.split(X, y)),
list(kf_randomized_iter_wrapped.split(X, y)))
splits_are_equal = True
except AssertionError:
splits_are_equal = False
assert_false(splits_are_equal, "If the splits are randomized, "
"successive calls to split should yield different results")
def test_imputation_copy():
"""Test imputation with copy=True."""
l = 5
# Test default behaviour and with copy=True
for params in [{}, {'copy': True}]:
X = sparse_random_matrix(l, l, density=0.75, random_state=0)
# Dense
imputer = Imputer(missing_values=0, strategy="mean", **params)
Xt = imputer.fit(X).transform(X)
Xt[0, 0] = np.nan
# Check that the objects are different and that they don't use
# the same buffer
assert_false(np.all(X.todense() == Xt))
# Sparse
imputer = Imputer(missing_values=0, strategy="mean", **params)
X = X.todense()
Xt = imputer.fit(X).transform(X)
Xt[0, 0] = np.nan
# Check that the objects are different and that they don't use
# the same buffer
assert_false(np.all(X == Xt))
def test_label_binarizer_set_label_encoding():
lb = LabelBinarizer(neg_label=-2, pos_label=0)
# two-class case with pos_label=0
inp = np.array([0, 1, 1, 0])
expected = np.array([[-2, 0, 0, -2]]).T
got = lb.fit_transform(inp)
assert_false(assert_warns(DeprecationWarning, getattr, lb, "multilabel_"))
assert_array_equal(expected, got)
assert_array_equal(lb.inverse_transform(got), inp)
lb = LabelBinarizer(neg_label=-2, pos_label=2)
# multi-class case
inp = np.array([3, 2, 1, 2, 0])
expected = np.array([[-2, -2, -2, +2],
[-2, -2, +2, -2],
[-2, +2, -2, -2],
[-2, -2, +2, -2],
[+2, -2, -2, -2]])
got = lb.fit_transform(inp)
assert_false(assert_warns(DeprecationWarning, getattr, lb, "multilabel_"))
assert_array_equal(expected, got)
assert_array_equal(lb.inverse_transform(got), inp)
def test_random_search_results():
# Make a dataset with a lot of noise to get various kind of prediction
# errors across CV folds and parameter settings
X, y = make_classification(n_samples=200, n_features=100, n_informative=3,
random_state=0)
# scipy.stats dists now supports `seed` but we still support scipy 0.12
# which doesn't support the seed. Hence the assertions in the test for
# random_search alone should not depend on randomization.
n_folds = 3
n_search_iter = 30
params = dict(C=expon(scale=10), gamma=expon(scale=0.1))
random_search = RandomizedSearchCV(SVC(), n_iter=n_search_iter, cv=n_folds,
iid=False, param_distributions=params)
random_search.fit(X, y)
random_search_iid = RandomizedSearchCV(SVC(), n_iter=n_search_iter,
cv=n_folds, iid=True,
param_distributions=params)
random_search_iid.fit(X, y)
param_keys = ('param_C', 'param_gamma')
score_keys = ('test_mean_score', 'test_rank_score',
'test_split0_score', 'test_split1_score',
'test_split2_score', 'test_std_score')
n_cand = n_search_iter
for search, iid in zip((random_search, random_search_iid), (False, True)):
assert_equal(iid, search.iid)
results = search.results_
# Check results structure
check_results_array_types(results, param_keys, score_keys)
check_results_keys(results, param_keys, score_keys, n_cand)
# For random_search, all the param array vals should be unmasked
assert_false(any(results['param_C'].mask) or
any(results['param_gamma'].mask))
check_results_grid_scores_consistency(search)
def test_column_transformer_get_set_params():
ct = ColumnTransformer([('trans1', StandardScaler(), [0]),
('trans2', StandardScaler(), [1])])
exp = {'n_jobs': None,
'remainder': 'drop',
'sparse_threshold': 0.3,
'trans1': ct.transformers[0][1],
'trans1__copy': True,
'trans1__with_mean': True,
'trans1__with_std': True,
'trans2': ct.transformers[1][1],
'trans2__copy': True,
'trans2__with_mean': True,
'trans2__with_std': True,
'transformers': ct.transformers,
'transformer_weights': None}
assert_dict_equal(ct.get_params(), exp)
ct.set_params(trans1__with_mean=False)
assert_false(ct.get_params()['trans1__with_mean'])
ct.set_params(trans1='passthrough')
exp = {'n_jobs': None,
'remainder': 'drop',
'sparse_threshold': 0.3,
'trans1': 'passthrough',
'trans2': ct.transformers[1][1],
'trans2__copy': True,
'trans2__with_mean': True,
'trans2__with_std': True,
'transformers': ct.transformers,
'transformer_weights': None}
assert_dict_equal(ct.get_params(), exp)
def test_equal_similarities_and_preferences():
# Unequal distances
X = np.array([[0, 0], [1, 1], [-2, -2]])
S = -euclidean_distances(X, squared=True)
assert_false(_equal_similarities_and_preferences(S, np.array(0)))
assert_false(_equal_similarities_and_preferences(S, np.array([0, 0])))
assert_false(_equal_similarities_and_preferences(S, np.array([0, 1])))
# Equal distances
X = np.array([[0, 0], [1, 1]])
S = -euclidean_distances(X, squared=True)
# Different preferences
assert_false(_equal_similarities_and_preferences(S, np.array([0, 1])))
# Same preferences
assert _equal_similarities_and_preferences(S, np.array([0, 0]))
assert _equal_similarities_and_preferences(S, np.array(0))
def test_check_is_permutation():
rng = np.random.RandomState(0)
p = np.arange(100)
rng.shuffle(p)
assert_true(_check_is_permutation(p, 100))
assert_false(_check_is_permutation(np.delete(p, 23), 100))
p[0] = 23
assert_false(_check_is_permutation(p, 100))
# Check if the additional duplicate indices are caught
assert_false(_check_is_permutation(np.hstack((p, 0)), 100))
def test_forest_attributes():
mr = MondrianForestRegressor(n_estimators=5, random_state=0)
mr.fit([[1, 2, 3], [4, 5, 6]], [1, 2])
assert_false(hasattr(mr, "classes_"))
assert_false(hasattr(mr, "n_classes_"))
mr.partial_fit([[1, 2, 3], [4, 5, 6]], [1, 2])
assert_false(hasattr(mr, "classes_"))
assert_false(hasattr(mr, "n_classes_"))
mr = MondrianForestClassifier(n_estimators=5, random_state=0)
mr.fit([[1, 2, 3], [4, 5, 6]], [1, 2])
assert_true(hasattr(mr, "classes_"))
assert_true(hasattr(mr, "n_classes_"))
mr = MondrianForestClassifier(n_estimators=5, random_state=0)
mr.partial_fit([[1, 2, 3], [4, 5, 6]], [1, 2])
assert_true(hasattr(mr, "classes_"))
assert_true(hasattr(mr, "n_classes_"))
def test_single_validate_constraint():
# Test categorical
cons = Single(0,1.0,'categorical')
assert_true(cons._validate_constraint(1.0))
assert_false(cons._validate_constraint(1.1))
cons = Single(0,'a','categorical')
assert_true(cons._validate_constraint('a'))
assert_false(cons._validate_constraint('b'))
# Test real
cons = Single(0,1.0,'real')
assert_true(cons._validate_constraint(1.0))
assert_false(cons._validate_constraint(1.1))
# Test integer
cons = Single(0,1,'integer')
assert_true(cons._validate_constraint(1))
assert_false(cons._validate_constraint(2))
def test_ovr_multilabel_predict_proba():
base_clf = MultinomialNB(alpha=1)
for au in (False, True):
X, Y = datasets.make_multilabel_classification(n_samples=100,
n_features=20,
n_classes=5,
n_labels=3,
length=50,
allow_unlabeled=au,
random_state=0)
X_train, Y_train = X[:80], Y[:80]
X_test = X[80:]
clf = OneVsRestClassifier(base_clf).fit(X_train, Y_train)
# Decision function only estimator.
decision_only = OneVsRestClassifier(svm.SVR(gamma='scale')
).fit(X_train, Y_train)
assert_false(hasattr(decision_only, 'predict_proba'))
# Estimator with predict_proba disabled, depending on parameters.
decision_only = OneVsRestClassifier(svm.SVC(gamma='scale',
probability=False))
assert_false(hasattr(decision_only, 'predict_proba'))
decision_only.fit(X_train, Y_train)
assert_false(hasattr(decision_only, 'predict_proba'))
assert_true(hasattr(decision_only, 'decision_function'))
# Estimator which can get predict_proba enabled after fitting
gs = GridSearchCV(svm.SVC(gamma='scale', probability=False),
param_grid={'probability': [True]})
proba_after_fit = OneVsRestClassifier(gs)
assert_false(hasattr(proba_after_fit, 'predict_proba'))
proba_after_fit.fit(X_train, Y_train)
assert_true(hasattr(proba_after_fit, 'predict_proba'))
Y_pred = clf.predict(X_test)
Y_proba = clf.predict_proba(X_test)
# predict assigns a label if the probability that the
# sample has the label is greater than 0.5.
pred = Y_proba > .5
assert_array_equal(pred, Y_pred)
def test_scaler_without_centering():
rng = np.random.RandomState(42)
X = rng.randn(4, 5)
X[:, 0] = 0.0 # first feature is always of zero
X_csr = sp.csr_matrix(X)
X_csc = sp.csc_matrix(X)
scaler = StandardScaler(with_mean=False).fit(X)
X_scaled = scaler.transform(X, copy=True)
assert_false(np.any(np.isnan(X_scaled)))
scaler_csr = StandardScaler(with_mean=False).fit(X_csr)
X_csr_scaled = scaler_csr.transform(X_csr, copy=True)
assert_false(np.any(np.isnan(X_csr_scaled.data)))
scaler_csc = StandardScaler(with_mean=False).fit(X_csc)
X_csc_scaled = scaler_csr.transform(X_csc, copy=True)
assert_false(np.any(np.isnan(X_csc_scaled.data)))
assert_equal(scaler.mean_, scaler_csr.mean_)
assert_array_almost_equal(scaler.std_, scaler_csr.std_)
assert_equal(scaler.mean_, scaler_csc.mean_)
assert_array_almost_equal(scaler.std_, scaler_csc.std_)
assert_array_almost_equal(
X_scaled.mean(axis=0), [0., -0.01, 2.24, -0.35, -0.78], 2)
assert_array_almost_equal(X_scaled.std(axis=0), [0., 1., 1., 1., 1.])
X_csr_scaled_mean, X_csr_scaled_std = mean_variance_axis0(X_csr_scaled)
assert_array_almost_equal(X_csr_scaled_mean, X_scaled.mean(axis=0))
assert_array_almost_equal(X_csr_scaled_std, X_scaled.std(axis=0))
# Check that X has not been modified (copy)
assert_true(X_scaled is not X)
assert_true(X_csr_scaled is not X_csr)
X_scaled_back = scaler.inverse_transform(X_scaled)
assert_true(X_scaled_back is not X)
assert_true(X_scaled_back is not X_scaled)
assert_array_almost_equal(X_scaled_back, X)
X_csr_scaled_back = scaler_csr.inverse_transform(X_csr_scaled)
assert_true(X_csr_scaled_back is not X_csr)
assert_true(X_csr_scaled_back is not X_csr_scaled)
assert_array_almost_equal(X_csr_scaled_back.toarray(), X)
X_csc_scaled_back = scaler_csr.inverse_transform(X_csc_scaled.tocsc())
assert_true(X_csc_scaled_back is not X_csc)
assert_true(X_csc_scaled_back is not X_csc_scaled)
assert_array_almost_equal(X_csc_scaled_back.toarray(), X)
def test_imputation_copy():
# Test imputation with copy
X_orig = sparse_random_matrix(5, 5, density=0.75, random_state=0)
# copy=True, dense => copy
X = X_orig.copy().toarray()
imputer = SimpleImputer(missing_values=0, strategy="mean", copy=True)
Xt = imputer.fit(X).transform(X)
Xt[0, 0] = -1
assert_false(np.all(X == Xt))
# copy=True, sparse csr => copy
X = X_orig.copy()
imputer = SimpleImputer(missing_values=X.data[0], strategy="mean",
copy=True)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_false(np.all(X.data == Xt.data))
# copy=False, dense => no copy
X = X_orig.copy().toarray()
imputer = SimpleImputer(missing_values=0, strategy="mean", copy=False)
Xt = imputer.fit(X).transform(X)
Xt[0, 0] = -1
assert_array_almost_equal(X, Xt)
# copy=False, sparse csr, axis=1 => no copy
X = X_orig.copy()
imputer = SimpleImputer(missing_values=X.data[0], strategy="mean",
copy=False, axis=1)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_array_almost_equal(X.data, Xt.data)
# copy=False, sparse csc => no copy
X = X_orig.copy().tocsc()
imputer = SimpleImputer(missing_values=X.data[0], strategy="mean",
copy=False)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_array_almost_equal(X.data, Xt.data)
# copy=False, sparse csr => copy
X = X_orig.copy()
imputer = SimpleImputer(missing_values=X.data[0], strategy="mean",
copy=False)
Xt = imputer.fit(X).transform(X)
Xt.data[0] = -1
assert_false(np.all(X.data == Xt.data))
def test_label_binarizer_column_y():
# first for binary classification vs multi-label with 1 possible class
# lists are multi-label, array is multi-class :-/
inp_list = [[1], [2], [1]]
inp_array = np.array(inp_list)
multilabel_indicator = np.array([[1, 0], [0, 1], [1, 0]])
binaryclass_array = np.array([[0], [1], [0]])
lb_1 = LabelBinarizer()
out_1 = lb_1.fit_transform(inp_list)
lb_2 = LabelBinarizer()
out_2 = lb_2.fit_transform(inp_array)
assert_array_equal(out_1, multilabel_indicator)
assert_true(assert_warns(DeprecationWarning, getattr, lb_1, "multilabel_"))
assert_false(assert_warns(DeprecationWarning, getattr, lb_1,
"indicator_matrix_"))
assert_array_equal(out_2, binaryclass_array)
assert_false(assert_warns(DeprecationWarning, getattr, lb_2,
"multilabel_"))
# second for multiclass classification vs multi-label with multiple
# classes
inp_list = [[1], [2], [1], [3]]
inp_array = np.array(inp_list)
# the indicator matrix output is the same in this case
indicator = np.array([[1, 0, 0], [0, 1, 0], [1, 0, 0], [0, 0, 1]])
lb_1 = LabelBinarizer()
out_1 = lb_1.fit_transform(inp_list)
lb_2 = LabelBinarizer()
out_2 = lb_2.fit_transform(inp_array)
assert_array_equal(out_1, out_2)
assert_true(assert_warns(DeprecationWarning, getattr, lb_1, "multilabel_"))
assert_array_equal(out_2, indicator)
assert_false(assert_warns(DeprecationWarning, getattr, lb_2,
"multilabel_"))
def test_bad_input():
# Test that it gives proper exception on deficient input
# impossible value of C
assert_raises(ValueError, svm.SVC(C=-1).fit, X, Y)
# impossible value of nu
clf = svm.NuSVC(nu=0.0)
assert_raises(ValueError, clf.fit, X, Y)
Y2 = Y[:-1] # wrong dimensions for labels
assert_raises(ValueError, clf.fit, X, Y2)
# Test with arrays that are non-contiguous.
for clf in (svm.SVC(), svm.LinearSVC(random_state=0)):
Xf = np.asfortranarray(X)
assert_false(Xf.flags['C_CONTIGUOUS'])
yf = np.ascontiguousarray(np.tile(Y, (2, 1)).T)
yf = yf[:, -1]
assert_false(yf.flags['F_CONTIGUOUS'])
assert_false(yf.flags['C_CONTIGUOUS'])
clf.fit(Xf, yf)
assert_array_equal(clf.predict(T), true_result)
# error for precomputed kernelsx
clf = svm.SVC(kernel='precomputed')
assert_raises(ValueError, clf.fit, X, Y)
# sample_weight bad dimensions
clf = svm.SVC()
assert_raises(ValueError, clf.fit, X, Y, sample_weight=range(len(X) - 1))
# predict with sparse input when trained with dense
clf = svm.SVC().fit(X, Y)
assert_raises(ValueError, clf.predict, sparse.lil_matrix(X))
Xt = np.array(X).T
clf.fit(np.dot(X, Xt), Y)
assert_raises(ValueError, clf.predict, X)
clf = svm.SVC()
clf.fit(X, Y)
assert_raises(ValueError, clf.predict, Xt)
def test_astype_copy_memory():
a_int32 = np.ones(3, np.int32)
# Check that dtype conversion works
b_float32 = astype(a_int32, dtype=np.float32, copy=False)
assert_equal(b_float32.dtype, np.float32)
# Changing dtype forces a copy even if copy=False
assert_false(np.may_share_memory(b_float32, a_int32))
# Check that copy can be skipped if requested dtype match
c_int32 = astype(a_int32, dtype=np.int32, copy=False)
assert_true(c_int32 is a_int32)
# Check that copy can be forced, and is the case by default:
d_int32 = astype(a_int32, dtype=np.int32, copy=True)
assert_false(np.may_share_memory(d_int32, a_int32))
e_int32 = astype(a_int32, dtype=np.int32)
assert_false(np.may_share_memory(e_int32, a_int32))
def test_label_binarizer():
lb = LabelBinarizer()
# one-class case defaults to negative label
inp = ["pos", "pos", "pos", "pos"]
expected = np.array([[0, 0, 0, 0]]).T
got = lb.fit_transform(inp)
assert_false(assert_warns(DeprecationWarning, getattr, lb, "multilabel_"))
assert_array_equal(lb.classes_, ["pos"])
assert_array_equal(expected, got)
assert_array_equal(lb.inverse_transform(got), inp)
# two-class case
inp = ["neg", "pos", "pos", "neg"]
expected = np.array([[0, 1, 1, 0]]).T
got = lb.fit_transform(inp)
assert_false(assert_warns(DeprecationWarning, getattr, lb, "multilabel_"))
assert_array_equal(lb.classes_, ["neg", "pos"])
assert_array_equal(expected, got)
to_invert = np.array([[1, 0],
[0, 1],
[0, 1],
[1, 0]])
assert_array_equal(lb.inverse_transform(to_invert), inp)
# multi-class case
inp = ["spam", "ham", "eggs", "ham", "0"]
expected = np.array([[0, 0, 0, 1],
[0, 0, 1, 0],
[0, 1, 0, 0],
[0, 0, 1, 0],
[1, 0, 0, 0]])
got = lb.fit_transform(inp)
assert_array_equal(lb.classes_, ['0', 'eggs', 'ham', 'spam'])
assert_false(assert_warns(DeprecationWarning, getattr, lb, "multilabel_"))
assert_array_equal(expected, got)
assert_array_equal(lb.inverse_transform(got), inp)
def test_metaestimator_delegation():
# Ensures specified metaestimators have methods iff subestimator does
def hides(method):
@property
def wrapper(obj):
if obj.hidden_method == method.__name__:
raise AttributeError('%r is hidden' % obj.hidden_method)
return functools.partial(method, obj)
return wrapper
class SubEstimator(BaseEstimator):
def __init__(self, param=1, hidden_method=None):
self.param = param
self.hidden_method = hidden_method
def fit(self, X, y=None, *args, **kwargs):
self.coef_ = np.arange(X.shape[1])
return True
def _check_fit(self):
check_is_fitted(self, 'coef_')
@hides
def inverse_transform(self, X, *args, **kwargs):
self._check_fit()
return X
@hides
def transform(self, X, *args, **kwargs):
self._check_fit()
return X
@hides
def predict(self, X, *args, **kwargs):
self._check_fit()
return np.ones(X.shape[0])
@hides
def predict_proba(self, X, *args, **kwargs):
self._check_fit()
return np.ones(X.shape[0])
@hides
def predict_log_proba(self, X, *args, **kwargs):
self._check_fit()
return np.ones(X.shape[0])
@hides
def decision_function(self, X, *args, **kwargs):
self._check_fit()
return np.ones(X.shape[0])
@hides
def score(self, X, *args, **kwargs):
self._check_fit()
return 1.0
methods = [
k for k in iterkeys(SubEstimator.__dict__)
if not k.startswith('_') and not k.startswith('fit')
]
methods.sort()
for delegator_data in DELEGATING_METAESTIMATORS:
delegate = SubEstimator()
delegator = delegator_data.construct(delegate)
for method in methods:
if method in delegator_data.skip_methods:
continue
assert_true(hasattr(delegate, method))
assert_true(
hasattr(delegator, method),
msg="%s does not have method %r when its delegate does" %
(delegator_data.name, method))
# delegation before fit raises a NotFittedError
assert_raises(NotFittedError, getattr(delegator, method),
delegator_data.fit_args[0])
delegator.fit(*delegator_data.fit_args)
for method in methods:
if method in delegator_data.skip_methods:
continue
# smoke test delegation
getattr(delegator, method)(delegator_data.fit_args[0])
for method in methods:
if method in delegator_data.skip_methods:
continue
delegate = SubEstimator(hidden_method=method)
delegator = delegator_data.construct(delegate)
assert_false(hasattr(delegate, method))
assert_false(hasattr(delegator, method),
msg="%s has method %r when its delegate does not" %
(delegator_data.name, method))
def test_all_estimator_no_base_class():
# test that all_estimators doesn't find abstract classes.
for name, Estimator in all_estimators():
msg = ("Base estimators such as {0} should not be included"
" in all_estimators").format(name)
assert_false(name.lower().startswith('base'), msg=msg)
def test_unfitted():
"""Non-regression: before fit, there should be not fitted attributes."""
ms = MeanShift()
assert_false(hasattr(ms, "cluster_centers_"))
assert_false(hasattr(ms, "labels_"))
def test_search_iid_param():
# Test the IID parameter
# noise-free simple 2d-data
X, y = make_blobs(centers=[[0, 0], [1, 0], [0, 1], [1, 1]],
random_state=0,
cluster_std=0.1,
shuffle=False,
n_samples=80)
# split dataset into two folds that are not iid
# first one contains data of all 4 blobs, second only from two.
mask = np.ones(X.shape[0], dtype=np.bool)
mask[np.where(y == 1)[0][::2]] = 0
mask[np.where(y == 2)[0][::2]] = 0
# this leads to perfect classification on one fold and a score of 1/3 on
# the other
# create "cv" for splits
cv = [[mask, ~mask], [~mask, mask]]
# once with iid=True (default)
grid_search = GridSearchCV(SVC(), param_grid={'C': [1, 10]}, cv=cv)
random_search = RandomizedSearchCV(SVC(),
n_iter=2,
param_distributions={'C': [1, 10]},
cv=cv)
for search in (grid_search, random_search):
search.fit(X, y)
assert_true(search.iid)
# Test the first candidate
cv_scores = np.array(
list(search.cv_results_['split%d_test_score' % s][0]
for s in range(search.n_splits_)))
mean = search.cv_results_['mean_test_score'][0]
std = search.cv_results_['std_test_score'][0]
assert_equal(search.cv_results_['param_C'][0], 1)
assert_array_almost_equal(cv_scores, [1, 1. / 3.])
# for first split, 1/4 of dataset is in test, for second 3/4.
# take weighted average and weighted std
expected_mean = 1 * 1. / 4. + 1. / 3. * 3. / 4.
expected_std = np.sqrt(1. / 4 * (expected_mean - 1)**2 + 3. / 4 *
(expected_mean - 1. / 3.)**2)
assert_almost_equal(mean, expected_mean)
assert_almost_equal(std, expected_std)
# once with iid=False
grid_search = GridSearchCV(SVC(),
param_grid={'C': [1, 10]},
cv=cv,
iid=False)
random_search = RandomizedSearchCV(SVC(),
n_iter=2,
param_distributions={'C': [1, 10]},
cv=cv,
iid=False)
for search in (grid_search, random_search):
search.fit(X, y)
assert_false(search.iid)
cv_scores = np.array(
list(search.cv_results_['split%d_test_score' % s][0]
for s in range(search.n_splits_)))
mean = search.cv_results_['mean_test_score'][0]
std = search.cv_results_['std_test_score'][0]
assert_equal(search.cv_results_['param_C'][0], 1)
# scores are the same as above
assert_array_almost_equal(cv_scores, [1, 1. / 3.])
# Unweighted mean/std is used
assert_almost_equal(mean, np.mean(cv_scores))
assert_almost_equal(std, np.std(cv_scores))
def test_pipeline_ducktyping():
pipeline = make_pipeline(Mult(5))
pipeline.predict
pipeline.transform
pipeline.inverse_transform
pipeline = make_pipeline(Transf())
assert_false(hasattr(pipeline, 'predict'))
pipeline.transform
pipeline.inverse_transform
pipeline = make_pipeline(None)
assert_false(hasattr(pipeline, 'predict'))
pipeline.transform
pipeline.inverse_transform
pipeline = make_pipeline(Transf(), NoInvTransf())
assert_false(hasattr(pipeline, 'predict'))
pipeline.transform
assert_false(hasattr(pipeline, 'inverse_transform'))
pipeline = make_pipeline(NoInvTransf(), Transf())
assert_false(hasattr(pipeline, 'predict'))
pipeline.transform
assert_false(hasattr(pipeline, 'inverse_transform'))
def test_pipeline_init():
# Test the various init parameters of the pipeline.
assert_raises(TypeError, Pipeline)
# Check that we can't instantiate pipelines with objects without fit
# method
assert_raises_regex(
TypeError, 'Last step of Pipeline should implement fit. '
'.*NoFit.*', Pipeline, [('clf', NoFit())])
# Smoke test with only an estimator
clf = NoTrans()
pipe = Pipeline([('svc', clf)])
assert_equal(
pipe.get_params(deep=True),
dict(svc__a=None, svc__b=None, svc=clf, **pipe.get_params(deep=False)))
# Check that params are set
pipe.set_params(svc__a=0.1)
assert_equal(clf.a, 0.1)
assert_equal(clf.b, None)
# Smoke test the repr:
repr(pipe)
# Test with two objects
clf = SVC()
filter1 = SelectKBest(f_classif)
pipe = Pipeline([('anova', filter1), ('svc', clf)])
# Check that we can't instantiate with non-transformers on the way
# Note that NoTrans implements fit, but not transform
assert_raises_regex(
TypeError, 'All intermediate steps should be transformers'
'.*\\bNoTrans\\b.*', Pipeline, [('t', NoTrans()), ('svc', clf)])
# Check that params are set
pipe.set_params(svc__C=0.1)
assert_equal(clf.C, 0.1)
# Smoke test the repr:
repr(pipe)
# Check that params are not set when naming them wrong
assert_raises(ValueError, pipe.set_params, anova__C=0.1)
# Test clone
pipe2 = assert_no_warnings(clone, pipe)
assert_false(pipe.named_steps['svc'] is pipe2.named_steps['svc'])
# Check that apart from estimators, the parameters are the same
params = pipe.get_params(deep=True)
params2 = pipe2.get_params(deep=True)
for x in pipe.get_params(deep=False):
params.pop(x)
for x in pipe2.get_params(deep=False):
params2.pop(x)
# Remove estimators that where copied
params.pop('svc')
params.pop('anova')
params2.pop('svc')
params2.pop('anova')
assert_equal(params, params2)
def test_sgd_proba(self):
# Check SGD.predict_proba
# Hinge loss does not allow for conditional prob estimate.
# We cannot use the factory here, because it defines predict_proba
# anyway.
clf = SGDClassifier(loss="hinge", alpha=0.01, n_iter=10).fit(X, Y)
assert_false(hasattr(clf, "predict_proba"))
assert_false(hasattr(clf, "predict_log_proba"))
# log and modified_huber losses can output probability estimates
# binary case
for loss in ["log", "modified_huber"]:
clf = self.factory(loss="modified_huber", alpha=0.01, n_iter=10)
clf.fit(X, Y)
p = clf.predict_proba([[3, 2]])
assert_true(p[0, 1] > 0.5)
p = clf.predict_proba([[-1, -1]])
assert_true(p[0, 1] < 0.5)
p = clf.predict_log_proba([[3, 2]])
assert_true(p[0, 1] > p[0, 0])
p = clf.predict_log_proba([[-1, -1]])
assert_true(p[0, 1] < p[0, 0])
# log loss multiclass probability estimates
clf = self.factory(loss="log", alpha=0.01, n_iter=10).fit(X2, Y2)
d = clf.decision_function([[.1, -.1], [.3, .2]])
p = clf.predict_proba([[.1, -.1], [.3, .2]])
assert_array_equal(np.argmax(p, axis=1), np.argmax(d, axis=1))
assert_almost_equal(p[0].sum(), 1)
assert_true(np.all(p[0] >= 0))
p = clf.predict_proba([[-1, -1]])
d = clf.decision_function([[-1, -1]])
assert_array_equal(np.argsort(p[0]), np.argsort(d[0]))
l = clf.predict_log_proba([[3, 2]])
p = clf.predict_proba([[3, 2]])
assert_array_almost_equal(np.log(p), l)
l = clf.predict_log_proba([[-1, -1]])
p = clf.predict_proba([[-1, -1]])
assert_array_almost_equal(np.log(p), l)
# Modified Huber multiclass probability estimates; requires a separate
# test because the hard zero/one probabilities may destroy the
# ordering present in decision_function output.
clf = self.factory(loss="modified_huber", alpha=0.01, n_iter=10)
clf.fit(X2, Y2)
d = clf.decision_function([[3, 2]])
p = clf.predict_proba([[3, 2]])
if not isinstance(self, SparseSGDClassifierTestCase):
assert_equal(np.argmax(d, axis=1), np.argmax(p, axis=1))
else: # XXX the sparse test gets a different X2 (?)
assert_equal(np.argmin(d, axis=1), np.argmin(p, axis=1))
# the following sample produces decision_function values < -1,
# which would cause naive normalization to fail (see comment
# in SGDClassifier.predict_proba)
x = X.mean(axis=0)
d = clf.decision_function([x])
if np.all(d < -1): # XXX not true in sparse test case (why?)
p = clf.predict_proba([x])
assert_array_almost_equal(p[0], [1 / 3.] * 3)
def test_scaler_int():
# test that scaler converts integer input to floating
# for both sparse and dense matrices
rng = np.random.RandomState(42)
X = rng.randint(20, size=(4, 5))
X[:, 0] = 0 # first feature is always of zero
X_csr = sparse.csr_matrix(X)
X_csc = sparse.csc_matrix(X)
null_transform = StandardScaler(with_mean=False, with_std=False, copy=True)
with warnings.catch_warnings(record=True):
X_null = null_transform.fit_transform(X_csr)
assert_array_equal(X_null.data, X_csr.data)
X_orig = null_transform.inverse_transform(X_null)
assert_array_equal(X_orig.data, X_csr.data)
with warnings.catch_warnings(record=True):
scaler = StandardScaler(with_mean=False).fit(X)
X_scaled = scaler.transform(X, copy=True)
assert_false(np.any(np.isnan(X_scaled)))
with warnings.catch_warnings(record=True):
scaler_csr = StandardScaler(with_mean=False).fit(X_csr)
X_csr_scaled = scaler_csr.transform(X_csr, copy=True)
assert_false(np.any(np.isnan(X_csr_scaled.data)))
with warnings.catch_warnings(record=True):
scaler_csc = StandardScaler(with_mean=False).fit(X_csc)
X_csc_scaled = scaler_csr.transform(X_csc, copy=True)
assert_false(np.any(np.isnan(X_csc_scaled.data)))
assert_equal(scaler.mean_, scaler_csr.mean_)
assert_array_almost_equal(scaler.std_, scaler_csr.std_)
assert_equal(scaler.mean_, scaler_csc.mean_)
assert_array_almost_equal(scaler.std_, scaler_csc.std_)
assert_array_almost_equal(X_scaled.mean(axis=0),
[0., 1.109, 1.856, 21., 1.559], 2)
assert_array_almost_equal(X_scaled.std(axis=0), [0., 1., 1., 1., 1.])
X_csr_scaled_mean, X_csr_scaled_std = mean_variance_axis0(
X_csr_scaled.astype(np.float))
assert_array_almost_equal(X_csr_scaled_mean, X_scaled.mean(axis=0))
assert_array_almost_equal(X_csr_scaled_std, X_scaled.std(axis=0))
# Check that X has not been modified (copy)
assert_true(X_scaled is not X)
assert_true(X_csr_scaled is not X_csr)
X_scaled_back = scaler.inverse_transform(X_scaled)
assert_true(X_scaled_back is not X)
assert_true(X_scaled_back is not X_scaled)
assert_array_almost_equal(X_scaled_back, X)
X_csr_scaled_back = scaler_csr.inverse_transform(X_csr_scaled)
assert_true(X_csr_scaled_back is not X_csr)
assert_true(X_csr_scaled_back is not X_csr_scaled)
assert_array_almost_equal(X_csr_scaled_back.toarray(), X)
X_csc_scaled_back = scaler_csr.inverse_transform(X_csc_scaled.tocsc())
assert_true(X_csc_scaled_back is not X_csc)
assert_true(X_csc_scaled_back is not X_csc_scaled)
assert_array_almost_equal(X_csc_scaled_back.toarray(), X)
def test_initialize_nn_output():
"""Test that NNDSVD does not return negative values"""
data = np.abs(random_state.randn(10, 10))
for var in (None, 'a', 'ar'):
W, H = nmf._initialize_nmf(data, 10)
assert_false((W < 0).any() or (H < 0).any())
def test_nls_nn_output():
"""Test that NLS solver doesn't return negative values"""
A = np.atleast_2d(range(1, 5))
Ap, _, _ = nmf._nls_subproblem(np.dot(A.T, -A), A.T, A, 0.001, 100)
assert_false((Ap < 0).any())
