def test_fit_max_chunk(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=20)
Nprobe, Nsample = (5, 4)
X = np.arange(Nprobe * Nsample).reshape((Nsample, Nprobe))
y = np.array(['A', 'A', 'B', 'B'])
with pytest.raises(ValueError):
dnet.fit(X, y)
def test_fit_transform(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.arange(Nprobe * Nsample).reshape((Nsample, Nprobe))
y = np.array(['A', 'A', 'B', 'B'])
Xnew = dnet.fit_transform(X, y)
assert Xnew.shape == (Nsample, 2)
assert np.allclose(Xnew, X[:, :2])
def test_set_signature(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.arange(Nprobe * Nsample).reshape((Nsample, Nprobe))
y = np.array(['A', 'A', 'B', 'B'])
dnet.fit(X, y)
with pytest.raises(ValueError):
signature = dnet.set_signature(-1)
def test_score(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.arange(Nprobe * Nsample).reshape((Nsample, Nprobe))
y = np.array(['A', 'A', 'B', 'B'])
dnet.fit(X, y)
score = dnet.score(X, y)
assert score == 0
def test_predict(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.full(shape=(Nsample, Nprobe), fill_value=42.)
# y = np.zeros(shape=(Nsample), dtype=int)
y = np.array(['A', 'A', 'B', 'B'])
with pytest.warns(RuntimeWarning):
dnet.fit(X, y)
y_pred = dnet.predict(X)
assert all(x == 0 for x in y_pred)
def test_random_fit(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.arange(Nprobe * Nsample).reshape((Nsample, Nprobe))
y = np.array(['A', 'A', 'B', 'B'])
dnet.fit(X, y)
signature = dnet.get_signature()[0]
assert all(x in ('number_of_genes', 'performace_couples', 'features',
'signature', 'score') for x in signature.keys())
assert signature['number_of_genes'] == 2
assert signature['score'] == 0
assert signature['performace_couples'] == 0
assert all(x == y for x, y in zip(signature['features'], (0, 1)))
def test_equal_fit(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.full(shape=(Nsample, Nprobe), fill_value=42.)
# y = np.zeros(shape=(Nsample), dtype=int)
y = np.array(['A', 'A', 'B', 'B'])
with pytest.warns(RuntimeWarning):
dnet.fit(X, y)
signature = dnet.signatures[0]
assert all(x in ('number_of_genes', 'performace_couples', 'features',
'signature', 'score') for x in signature.keys())
assert signature['number_of_genes'] == 2
assert signature['score'] == 0.5
assert signature['performace_couples'] == 2
assert all(x == y for x, y in zip(signature['features'], (0, 1)))
def test_estimator(self):
params = {
'estimator': None,
'cv': LeaveOneOut(),
'scoring': None,
'max_chunk': 1,
'percentage': .1,
'verbose': False,
'n_jobs': 1
}
with pytest.raises(ValueError):
dnet = DNetPRO(**params)
def test_n_jobs(self):
params = {
'estimator': GaussianNB(priors=None, var_smoothing=1e-09),
'cv': LeaveOneOut(),
'scoring': None,
'max_chunk': 10,
'percentage': .1,
'verbose': False,
'n_jobs': 0
}
with pytest.raises(ValueError):
dnet = DNetPRO(**params)
def test_constructor(self):
params = {
'estimator': GaussianNB(priors=None, var_smoothing=1e-09),
'cv': LeaveOneOut(),
'scoring': None,
'max_chunk': 100,
'percentage': .1,
'verbose': False,
'n_jobs': 1
}
dnet = DNetPRO(**params)
print(dnet)
def test_predict_logproba(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
Nprobe, Nsample = (5, 4)
X = np.full(shape=(Nsample, Nprobe), fill_value=42.)
# y = np.zeros(shape=(Nsample), dtype=int)
y = np.array(['A', 'A', 'B', 'B'])
with pytest.warns(RuntimeWarning):
dnet.fit(X, y)
proba = dnet.predict_log_proba(dnet.transform(X))
assert np.isnan(proba).sum() == np.prod(proba.shape)
def test_label2numbers(self):
y = ('A', 'A', 'B', 'B')
num_y = DNetPRO.label2numbers(y)
assert all(x == y for x, y in zip(num_y, (0, 0, 1, 1)))
def test_pendrem_star(self):
graph = nx.star_graph(5)
g = DNetPRO.pendrem(graph)
assert all(x == 0 for x in g.nodes())
n_clusters_per_class=1,
class_sep=1.,
scale=1.,
shuffle=False,
random_state=123)
classifier = GaussianNB()
# Split the dataset in training and test for performances evaluation
X_train, X_test, y_train, y_test = train_test_split(data,
label,
test_size=0.33,
random_state=42)
# Create the DNetPRO feature selection object with GaussianNB classifier
dnet = DNetPRO(estimator=classifier,
scoring='accuracy',
n_jobs=4,
verbose=False)
# extract the filtered dataset as the signature with highest score in the training set
Dnet_data = dnet.fit_transform(X_train, y_train)
new_sample, new_probe = Dnet_data.shape
# Best DNetPRO signature
dnet_signature = dnet.selected_signature
# print some informations
print('Signature DNetPRO: {}'.format(sorted(dnet_signature)))
print('DNetPRO score: {:.3f}'.format(dnet.score(X_test, y_test)))
print('Informative found: {:d} / {:d}'.format(
len([x for x in dnet_signature if x < Ninformative]), Ninformative))
# Compare the obtained results against the Kbest features with K=number of feature in the DNetPRO signature
def test_pendrem_path(self):
graph = nx.path_graph(5)
g = DNetPRO.pendrem(graph)
assert all(x == y for x, y in zip(g.nodes(), graph.nodes))
def test_is_fitted(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
with pytest.raises(NotFittedError):
pred_label = dnet.predict([])
assert pred_label is not None
def test_estimator_type(self):
dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4)
assert dnet._estimator_type == 'classifier'
def test_pendrem_full(self):
graph = nx.complete_graph(5)
g = DNetPRO.pendrem(graph)
assert all(x == y for x, y in zip(g.nodes(), graph.nodes))
n_clusters_per_class=1,
class_sep=1.,
scale=1.,
shuffle=False,
random_state=seed)
X_train, X_test, y_train, y_test = train_test_split(
data, label, test_size=0.33, random_state=42)
# dnet_tic = now()
# DNetPRO feature selection
dnet = DNetPRO(estimator=classifier,
scoring='accuracy',
n_jobs=available_cores,
verbose=False,
max_chunk=10)
# Dnet_data = dnet.fit_transform(X_train, y_train)
# new_sample, new_probe = Dnet_data.shape
# dnet_signature = dnet.selected_signature
# # DNetPRO parameters to save
# dnet_score = dnet.score(X_test, y_test)
# dnet_informative = len([x for x in dnet_signature if x < Ninformative])
# dnet_size = len(dnet_signature)
# dnet_toc = now()
#!/usr/bin/env python # -*- coding: utf-8 -*- import numpy as np from DNetPRO import DNetPRO from sklearn.naive_bayes import GaussianNB if __name__ == '__main__': Nprobe, Nsample = (5, 4) X = np.random.uniform(low=0., high=1., size=(Nsample, Nprobe)) y = np.array(['A', 'A', 'B', 'B']) dnet = DNetPRO(estimator=GaussianNB(), max_chunk=4) dnet.fit(X, y) print(dnet.signatures)