class GENDISFeatures(BaseEstimator, TransformerMixin):
def __init__(self):
pass
def fit(self, X, y):
print(X.shape)
self.genetic_extractor = GeneticExtractor(verbose=True,
population_size=25,
iterations=10,
wait=5,
max_len=50,
plot=None,
location=True,
n_jobs=4,
fitness=auc_fitness_location)
self.genetic_extractor.fit(X, y)
self.names = []
for i, shap in enumerate(self.genetic_extractor.shapelets):
self.names.append('dist_shap_{}'.format(i))
for i, shap in enumerate(self.genetic_extractor.shapelets):
self.names.append('loc_shap_{}'.format(i))
return self
def transform(self, X):
return self.genetic_extractor.transform(X)
def fit_transform(self, X, y):
self.fit(X, y)
return self.transform(X)
def test_accept_string_labels():
X = [
[0] * 8,
[0] * 8,
[0] * 8,
[0] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
]
y = ['a', 'a', 'a', 'a', 'b', 'b', 'b', 'b']
genetic = GeneticExtractor(population_size=5, iterations=5)
genetic.fit(X, y)
def test_accept_float_labels():
X = [
[0] * 8,
[0] * 8,
[0] * 8,
[0] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
]
y = [1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 2.0]
genetic = GeneticExtractor(population_size=5, iterations=5)
genetic.fit(X, y)
def test_accept_list():
X = [
[0] * 8,
[0] * 8,
[0] * 8,
[0] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
]
y = [0, 0, 0, 0, 1, 1, 1, 1]
genetic = GeneticExtractor(population_size=5, iterations=5)
genetic.fit(X, y)
def test_accept_variable_length_arrays():
X = [
[0] * 10,
[0] * 6,
[0] * 8,
[0] * 6,
[1] * 8,
[1] * 7,
[1] * 8,
[1] * 5,
]
y = [0, 0, 0, 0, 1, 1, 1, 1]
genetic = GeneticExtractor(population_size=5, iterations=5)
genetic.fit(X, y)
def test_accept_pd_DataFrame():
X = [
[0] * 8,
[0] * 8,
[0] * 8,
[0] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
]
y = [0, 0, 0, 0, 1, 1, 1, 1]
pd_X = pd.DataFrame(X)
pd_y = pd.Series(y)
genetic = GeneticExtractor(population_size=5, iterations=5)
genetic.fit(pd_X, pd_y)
def fit(self, X, y):
print(X.shape)
self.genetic_extractor = GeneticExtractor(verbose=True,
population_size=25,
iterations=10,
wait=5,
max_len=50,
plot=None,
location=True,
n_jobs=4,
fitness=auc_fitness_location)
self.genetic_extractor.fit(X, y)
self.names = []
for i, shap in enumerate(self.genetic_extractor.shapelets):
self.names.append('dist_shap_{}'.format(i))
for i, shap in enumerate(self.genetic_extractor.shapelets):
self.names.append('loc_shap_{}'.format(i))
return self
def teast_pipeline():
X, y = random_walk_blobs(n_ts_per_blob=20, sz=64, noise_level=0.1)
X = np.reshape(X, (X.shape[0], X.shape[1]))
extractor = GeneticExtractor(iterations=5, n_jobs=1, population_size=10)
lr = LogisticRegression()
pipeline = Pipeline([
('shapelets', extractor),
('log_reg', lr)
])
pipeline.fit(X, y)
def test_accept_np_array():
X = [
[0] * 8,
[0] * 8,
[0] * 8,
[0] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
[1] * 8,
]
y = [0, 0, 0, 0, 1, 1, 1, 1]
np_X = []
for x in X:
np_X.append(np.array(x))
np_X = np.array(np_X)
np_y = np.array(y)
genetic = GeneticExtractor(population_size=5, iterations=5)
genetic.fit(np_X, np_y)
def test_serialization():
X, y = random_walk_blobs(n_ts_per_blob=20, sz=64, noise_level=0.1)
X = np.reshape(X, (X.shape[0], X.shape[1]))
extractor = GeneticExtractor(iterations=5, n_jobs=1, population_size=10)
distances = extractor.fit_transform(X, y)
extractor.save('temp.p')
new_extractor = GeneticExtractor.load('temp.p')
new_distances = new_extractor.transform(X)
np.testing.assert_array_equal(distances, new_distances)
os.remove('temp.p')