def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = iris.target
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_true(X_r.shape == iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=3,
loss_func=zero_one)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_true(X_r.shape == iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
def test_number_of_subsets_of_features():
# In RFE, 'number_of_subsets_of_features'
# = the number of iterations in '_fit'
# = max(ranking_)
# = 1 + (n_features + step - n_features_to_select - 1) // step
# After optimization #4534, this number
# = 1 + np.ceil((n_features - n_features_to_select) / float(step))
# This test case is to test their equivalence, refer to #4534 and #3824
def formula1(n_features, n_features_to_select, step):
return 1 + ((n_features + step - n_features_to_select - 1) // step)
def formula2(n_features, n_features_to_select, step):
return 1 + np.ceil((n_features - n_features_to_select) / float(step))
# RFE
# Case 1, n_features - n_features_to_select is divisible by step
# Case 2, n_features - n_features_to_select is not divisible by step
n_features_list = [11, 11]
n_features_to_select_list = [3, 3]
step_list = [2, 3]
for n_features, n_features_to_select, step in zip(
n_features_list, n_features_to_select_list, step_list):
generator = check_random_state(43)
X = generator.normal(size=(100, n_features))
y = generator.rand(100).round()
rfe = RFE(estimator=SVC(kernel="linear"),
n_features_to_select=n_features_to_select,
step=step)
rfe.fit(X, y)
# this number also equals to the maximum of ranking_
assert_equal(np.max(rfe.ranking_),
formula1(n_features, n_features_to_select, step))
assert_equal(np.max(rfe.ranking_),
formula2(n_features, n_features_to_select, step))
# In RFECV, 'fit' calls 'RFE._fit'
# 'number_of_subsets_of_features' of RFE
# = the size of 'grid_scores' of RFECV
# = the number of iterations of the for loop before optimization #4534
# RFECV, n_features_to_select = 1
# Case 1, n_features - 1 is divisible by step
# Case 2, n_features - 1 is not divisible by step
n_features_to_select = 1
n_features_list = [11, 10]
step_list = [2, 2]
for n_features, step in zip(n_features_list, step_list):
generator = check_random_state(43)
X = generator.normal(size=(100, n_features))
y = generator.rand(100).round()
rfecv = RFECV(estimator=SVC(kernel="linear"), step=step, cv=5)
rfecv.fit(X, y)
assert_equal(rfecv.grid_scores_.shape[0],
formula1(n_features, n_features_to_select, step))
assert_equal(rfecv.grid_scores_.shape[0],
formula2(n_features, n_features_to_select, step))
def _fs_rfecv(self, data, labels, plot_filename, sample = 0.05):
"""
Helper function to perform feature selection with Recursive Feature Elimination based
on a cross-validation over the entire or part of the data set.
@param data: the values of the features
@type data: numpy.array
@param labels: the values of the training labels
@type labels: numpy.array
@param plot_filename: a string where the graph will be stored
@type plot_filename: string
@return: a trained transformer, the modified data and the co-efficients assigned to each feature
@rtype: tuple of (scikit.transformer, numpy.array, OrderedDict)
"""
attributes = OrderedDict()
svc = SVC(kernel="linear")
if sample:
skf = StratifiedKFold(labels, n_folds=int(round(1.00/sample)))
last_fold = list(skf)[-1]
sampledata = np.array([data[index] for index in last_fold[1]])
samplelabels = np.array([labels[index] for index in last_fold[1]])
log.info("RFECV will be performed on data: {}".format(sampledata.shape))
else:
sampledata = data
samplelabels = labels
transformer = RFECV(estimator=svc, step=1, cv=StratifiedKFold(samplelabels, 5),
scoring='accuracy')
log.info("scikit: Running feature selection RFECV_SVC")
log.info("scikit: data dimensions before fit_transform(): {}".format(data.shape))
log.info("scikit: labels dimensions before fit_transform(): {}".format(labels.shape))
transformer.fit(sampledata, samplelabels)
log.info("scikit: Data fit. Proceeding with transforming...")
data = transformer.transform(data)
log.info("scikit: Dimensions after fit_transform(): %s,%s" % data.shape)
#produce a plot if requested and supported (for RFE)
if plot_filename:
try:
grid_scores = transformer.grid_scores_
except:
return transformer, data, attributes
plt.figure()
plt.xlabel("Number of features selected")
plt.ylabel("Cross validation score (nb of correct classifications)")
plt.plot(range(1, len(grid_scores) + 1), grid_scores)
plt.savefig(plot_filename, bbox_inches='tight')
#put ranks in an array, so that we can get them in the log file
for i, rank_strings in enumerate(transformer.ranking_):
attributes["RFE_rank_f{}".format(i)] = rank_strings
for i, rank_strings in enumerate(transformer.support_):
attributes["RFE_mask_f{}".format(i)] = rank_strings
return transformer, data, attributes
def test_number_of_subsets_of_features():
# In RFE, 'number_of_subsets_of_features'
# = the number of iterations in '_fit'
# = max(ranking_)
# = 1 + (n_features + step - n_features_to_select - 1) // step
# After optimization #4534, this number
# = 1 + np.ceil((n_features - n_features_to_select) / float(step))
# This test case is to test their equivalence, refer to #4534 and #3824
def formula1(n_features, n_features_to_select, step):
return 1 + ((n_features + step - n_features_to_select - 1) // step)
def formula2(n_features, n_features_to_select, step):
return 1 + np.ceil((n_features - n_features_to_select) / float(step))
# RFE
# Case 1, n_features - n_features_to_select is divisible by step
# Case 2, n_features - n_features_to_select is not divisible by step
n_features_list = [11, 11]
n_features_to_select_list = [3, 3]
step_list = [2, 3]
for n_features, n_features_to_select, step in zip(
n_features_list, n_features_to_select_list, step_list):
generator = check_random_state(43)
X = generator.normal(size=(100, n_features))
y = generator.rand(100).round()
rfe = RFE(estimator=SVC(kernel="linear"),
n_features_to_select=n_features_to_select, step=step)
rfe.fit(X, y)
# this number also equals to the maximum of ranking_
assert_equal(np.max(rfe.ranking_),
formula1(n_features, n_features_to_select, step))
assert_equal(np.max(rfe.ranking_),
formula2(n_features, n_features_to_select, step))
# In RFECV, 'fit' calls 'RFE._fit'
# 'number_of_subsets_of_features' of RFE
# = the size of 'grid_scores' of RFECV
# = the number of iterations of the for loop before optimization #4534
# RFECV, n_features_to_select = 1
# Case 1, n_features - 1 is divisible by step
# Case 2, n_features - 1 is not divisible by step
n_features_to_select = 1
n_features_list = [11, 10]
step_list = [2, 2]
for n_features, step in zip(n_features_list, step_list):
generator = check_random_state(43)
X = generator.normal(size=(100, n_features))
y = generator.rand(100).round()
rfecv = RFECV(estimator=SVC(kernel="linear"), step=step, cv=5)
rfecv.fit(X, y)
assert_equal(rfecv.grid_scores_.shape[0],
formula1(n_features, n_features_to_select, step))
assert_equal(rfecv.grid_scores_.shape[0],
formula2(n_features, n_features_to_select, step))
def test_rfecv_mockclassifier():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=MockClassifier(), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
def test_rfecv_mockclassifier():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=MockClassifier(), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
def run_rfe1(lang_pair, results_fname, subset=None, step=100, folds=2):
ambig_fname = config["sample"][lang_pair]["ambig_fname"]
ambig_map = AmbiguityMap(ambig_fname, subset=subset)
samples_fname = config["sample"][lang_pair]["samples_filt_fname"]
sample_hdfile = h5py.File(samples_fname, "r")
data_gen = DataSetGenerator(ambig_map, sample_hdfile)
estimator = SGDClassifier()
descriptor = [ ("lemma", "U32"),
("pos", "U32"),
("#cand", "i"),
("#feats", "i"),
("prec", "f"),
("rec", "f"),
("f-score", "f")]
results = np.zeros(len(ambig_map), dtype=descriptor)
i = 0
for data in data_gen:
print i+1, data.source_lempos,
if not data.target_lempos:
print "*** no samples ***"
continue
lemma, pos = data.source_lempos.rsplit("/", 1)
n_cand = len(data.target_lempos)
samples = data.samples.tocsr()
# Fix scoring func
rfecv = RFECV(estimator=estimator,
step=step,
cv=StratifiedKFold(data.targets, folds),
loss_func=loss_func,
verbose=True
)
rfecv.fit(samples, data.targets)
samples = rfecv.transform(samples)
scores = cross_val_score(estimator,
samples,
data.targets,
score_func=score_func)
scores = scores.mean(axis=1) * 100
results[i] = (lemma, pos, n_cand, rfecv.n_features_) + tuple(scores)
print results[i]
print rfecv.cv_scores_
i += 1
np.save(results_fname, results[:i])
def test_rfe_cv_n_jobs():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = iris.target
rfecv = RFECV(estimator=SVC(kernel='linear'))
rfecv.fit(X, y)
rfecv_ranking = rfecv.ranking_
rfecv_grid_scores = rfecv.grid_scores_
rfecv.set_params(n_jobs=2)
rfecv.fit(X, y)
assert_array_almost_equal(rfecv.ranking_, rfecv_ranking)
assert_array_almost_equal(rfecv.grid_scores_, rfecv_grid_scores)
def test_rfe_cv_groups():
generator = check_random_state(0)
iris = load_iris()
number_groups = 4
groups = np.floor(np.linspace(0, number_groups, len(iris.target)))
X = iris.data
y = (iris.target > 0).astype(int)
est_groups = RFECV(
estimator=RandomForestClassifier(random_state=generator),
step=1,
scoring='accuracy',
cv=GroupKFold(n_splits=2))
est_groups.fit(X, y, groups=groups)
assert est_groups.n_features_ > 0
def test_rfe_cv_n_jobs():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = iris.target
rfecv = RFECV(estimator=SVC(kernel='linear'))
rfecv.fit(X, y)
rfecv_ranking = rfecv.ranking_
rfecv_grid_scores = rfecv.grid_scores_
rfecv.set_params(n_jobs=2)
rfecv.fit(X, y)
assert_array_almost_equal(rfecv.ranking_, rfecv_ranking)
assert_array_almost_equal(rfecv.grid_scores_, rfecv_grid_scores)
def test_rfe_cv_groups():
generator = check_random_state(0)
iris = load_iris()
number_groups = 4
groups = np.floor(np.linspace(0, number_groups, len(iris.target)))
X = iris.data
y = (iris.target > 0).astype(int)
est_groups = RFECV(
estimator=RandomForestClassifier(random_state=generator),
step=1,
scoring='accuracy',
cv=GroupKFold(n_splits=2)
)
est_groups.fit(X, y, groups=groups)
assert est_groups.n_features_ > 0
def test_rfecv_verbose_output():
# Check verbose=1 is producing an output.
from io import StringIO
import sys
sys.stdout = StringIO()
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target)
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5, verbose=1)
rfecv.fit(X, y)
verbose_output = sys.stdout
verbose_output.seek(0)
assert_greater(len(verbose_output.readline()), 0)
def test_rfecv_verbose_output():
# Check verbose=1 is producing an output.
from sklearn.externals.six.moves import cStringIO as StringIO
import sys
sys.stdout = StringIO()
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target)
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5, verbose=1)
rfecv.fit(X, y)
verbose_output = sys.stdout
verbose_output.seek(0)
assert_greater(len(verbose_output.readline()), 0)
def test_rfecv_grid_scores_size():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Non-regression test for varying combinations of step and
# min_features_to_select.
for step, min_features_to_select in [[2, 1], [2, 2], [3, 3]]:
rfecv = RFECV(estimator=MockClassifier(), step=step,
min_features_to_select=min_features_to_select, cv=5)
rfecv.fit(X, y)
score_len = np.ceil(
(X.shape[1] - min_features_to_select) / step) + 1
assert len(rfecv.grid_scores_) == score_len
assert len(rfecv.ranking_) == X.shape[1]
assert rfecv.n_features_ >= min_features_to_select
def test_rfecv_grid_scores_size():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Non-regression test for varying combinations of step and
# min_features_to_select.
for step, min_features_to_select in [[2, 1], [2, 2], [3, 3]]:
rfecv = RFECV(estimator=MockClassifier(), step=step,
min_features_to_select=min_features_to_select)
rfecv.fit(X, y)
score_len = np.ceil(
(X.shape[1] - min_features_to_select) / step) + 1
assert len(rfecv.grid_scores_) == score_len
assert len(rfecv.ranking_) == X.shape[1]
assert rfecv.n_features_ >= min_features_to_select
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# All the noisy variable were filtered out
assert_array_equal(X_r, iris.data)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Test using a customized loss function
scoring = make_scorer(zero_one_loss, greater_is_better=False)
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=5,
scoring=scoring)
ignore_warnings(rfecv.fit)(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test using a scorer
scorer = SCORERS['accuracy']
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5, scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test fix on grid_scores
def test_scorer(estimator, X, y):
return 1.0
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=5,
scoring=test_scorer)
rfecv.fit(X, y)
assert_array_equal(rfecv.grid_scores_, np.ones(len(rfecv.grid_scores_)))
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.cv_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
assert_array_almost_equal(X_r_sparse.toarray(), X_r)
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3,
loss_func=zero_one)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = iris.target
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.cv_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
assert_array_almost_equal(X_r_sparse.toarray(), X_r)
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=3,
loss_func=zero_one)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = iris.target
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
assert_array_almost_equal(X_r_sparse.toarray(), X_r)
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3,
loss_func=zero_one)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# All the noisy variable were filtered out
assert_array_equal(X_r, iris.data)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Test using a customized loss function
scoring = make_scorer(zero_one_loss, greater_is_better=False)
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=scoring)
ignore_warnings(rfecv.fit)(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test using a scorer
scorer = get_scorer('accuracy')
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test fix on grid_scores
def test_scorer(estimator, X, y):
return 1.0
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=test_scorer)
rfecv.fit(X, y)
assert_array_equal(rfecv.grid_scores_, np.ones(len(rfecv.grid_scores_)))
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = iris.target
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear", C=100), step=1, cv=3)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_true(X_r.shape == iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear", C=100), step=1, cv=3,
loss_func=zero_one)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_true(X_r.shape == iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
assert_array_almost_equal(X_r_sparse.toarray(), X_r)
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=3,
loss_func=zero_one_loss)
with warnings.catch_warnings(record=True):
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
# Test using a scorer
scorer = SCORERS['accuracy']
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=3, scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_equal(X_r.shape, iris.data.shape)
assert_array_almost_equal(X_r[:10], iris.data[:10])
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# All the noisy variable were filtered out
assert_array_equal(X_r, iris.data)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Test using a customized loss function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
loss_func=zero_one_loss)
with warnings.catch_warnings(record=True):
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test using a scorer
scorer = SCORERS['accuracy']
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
def test_rfecv(X, y):
# generator = check_random_state(0)
# iris = load_iris()
# X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
# y = list(iris.target) # regression test: list should be supported
# print(X)
# print(y)
# Test using the score function
scorer = get_scorer('accuracy')
rfecv = RFECV(estimator=LogisticRegression(),
step=1,
cv=10,
scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
print(rfecv.ranking_)
print(rfecv.support_)
print(X_r.shape)
# All the noisy variable were filtered out
#assert_array_equal(X_r, X)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
print(X_r_sparse.shape)
#assert_array_equal(X_r_sparse.toarray(), X)
# Test using a customized loss function
scoring = make_scorer(zero_one_loss, greater_is_better=False)
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=5,
scoring=scoring)
ignore_warnings(rfecv.fit)(X, y)
X_r = rfecv.transform(X)
print(X_r.shape)
#assert_array_equal(X_r, X)
# Test using a scorer
scorer = get_scorer('accuracy')
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5, scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
print(X_r.shape)
#assert_array_equal(X_r, X)
# Same as the first two tests, but with step=2
rfecv = RFECV(estimator=SVC(kernel="linear"), step=2, cv=5)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
print(X_r.shape)
#assert_array_equal(X_r, X)
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=2, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
print(X_r_sparse.shape)
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# All the noisy variable were filtered out
assert_array_equal(X_r, iris.data)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Test using a customized loss function
scoring = make_scorer(zero_one_loss, greater_is_better=False)
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=5,
scoring=scoring)
ignore_warnings(rfecv.fit)(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test using a scorer
scorer = get_scorer('accuracy')
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5, scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test fix on grid_scores
def test_scorer(estimator, X, y):
return 1.0
rfecv = RFECV(estimator=SVC(kernel="linear"),
step=1,
cv=5,
scoring=test_scorer)
rfecv.fit(X, y)
assert_array_equal(rfecv.grid_scores_, np.ones(len(rfecv.grid_scores_)))
# In the event of cross validation score ties, the expected behavior of
# RFECV is to return the FEWEST features that maximize the CV score.
# Because test_scorer always returns 1.0 in this example, RFECV should
# reduce the dimensionality to a single feature (i.e. n_features_ = 1)
assert_equal(rfecv.n_features_, 1)
# Same as the first two tests, but with step=2
rfecv = RFECV(estimator=SVC(kernel="linear"), step=2, cv=5)
rfecv.fit(X, y)
assert_equal(len(rfecv.grid_scores_), 6)
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=2, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Verifying that steps < 1 don't blow up.
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=.2, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
def test_rfecv():
generator = check_random_state(0)
iris = load_iris()
X = np.c_[iris.data, generator.normal(size=(len(iris.data), 6))]
y = list(iris.target) # regression test: list should be supported
# Test using the score function
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
rfecv.fit(X, y)
# non-regression test for missing worst feature:
assert_equal(len(rfecv.grid_scores_), X.shape[1])
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
# All the noisy variable were filtered out
assert_array_equal(X_r, iris.data)
# same in sparse
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Test using a customized loss function
scoring = make_scorer(zero_one_loss, greater_is_better=False)
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=scoring)
ignore_warnings(rfecv.fit)(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test using a scorer
scorer = get_scorer('accuracy')
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=scorer)
rfecv.fit(X, y)
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
# Test fix on grid_scores
def test_scorer(estimator, X, y):
return 1.0
rfecv = RFECV(estimator=SVC(kernel="linear"), step=1, cv=5,
scoring=test_scorer)
rfecv.fit(X, y)
assert_array_equal(rfecv.grid_scores_, np.ones(len(rfecv.grid_scores_)))
# In the event of cross validation score ties, the expected behavior of
# RFECV is to return the FEWEST features that maximize the CV score.
# Because test_scorer always returns 1.0 in this example, RFECV should
# reduce the dimensionality to a single feature (i.e. n_features_ = 1)
assert_equal(rfecv.n_features_, 1)
# Same as the first two tests, but with step=2
rfecv = RFECV(estimator=SVC(kernel="linear"), step=2, cv=5)
rfecv.fit(X, y)
assert_equal(len(rfecv.grid_scores_), 6)
assert_equal(len(rfecv.ranking_), X.shape[1])
X_r = rfecv.transform(X)
assert_array_equal(X_r, iris.data)
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=2, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)
# Verifying that steps < 1 don't blow up.
rfecv_sparse = RFECV(estimator=SVC(kernel="linear"), step=.2, cv=5)
X_sparse = sparse.csr_matrix(X)
rfecv_sparse.fit(X_sparse, y)
X_r_sparse = rfecv_sparse.transform(X_sparse)
assert_array_equal(X_r_sparse.toarray(), iris.data)