def main():
hsic_lasso = HSICLasso()
hsic_lasso.input("../tests/test_data/csv_data_mv.csv",
output_list=['output1', 'output2'])
hsic_lasso.regression(5)
hsic_lasso.dump()
hsic_lasso.plot_path()
def hsic_sel(csv, no_features, method='classification'):
hsic_lasso = HSICLasso()
hsic_lasso.input(csv)
if method == 'regression':
hsic_lasso.regression(no_features)
else:
hsic_lasso.classification(no_features)
return hsic_lasso.get_features()
def hsic(num_features, hsic_data, method='regression'):
hsic_lasso = HSICLasso()
hsic_lasso.input(hsic_data)
if method == 'regression':
hsic_lasso.regression(num_features)
else:
hsic_lasso.classification(num_features)
return hsic_lasso.get_features()
def main():
hsic_lasso = HSICLasso()
hsic_lasso.input("../tests/test_data/matlab_data.mat")
#max_neighbors=0 means that we only use the HSIC Lasso features to plot heatmap
hsic_lasso.regression(5, max_neighbors=0)
#Compute linkage
hsic_lasso.linkage()
#Run Hierarchical clustering
# Features are clustered by using HSIC scores
# Samples are clusterd by using Euclid distance
hsic_lasso.plot_heatmap()
def main():
hsic_lasso = HSICLasso()
hsic_lasso.input("../tests/test_data/matlab_data.mat")
#Single core processing
hsic_lasso.regression(5, n_jobs=1)
#Multi-core processing. Use all available cores (default)
hsic_lasso.regression(5, n_jobs=-1)
def fit(self, X, y):
sss = StratifiedShuffleSplit(n_splits=self.hsic_splits,
random_state=42)
idxs = []
hsics = []
for train_index, test_index in list(sss.split(X, y)):
hsic_lasso2 = HSICLasso()
hsic_lasso2.input(X[train_index], y[train_index])
hsic_lasso2.classification(
self.n_features, B=self.B,
M=self.M) #(self.n_features, B=self.B, M=self.M)
hsics.append(hsic_lasso2)
# not just best features - get their neighbors (similar features) too
all_ft_idx = np.array(hsic_lasso2.get_index(), dtype=int).ravel()
for i in range(len(all_ft_idx)):
idx = np.array(hsic_lasso2.get_index_neighbors(
feat_index=i, num_neighbors=10),
dtype=int)
score = np.array(hsic_lasso2.get_index_neighbors_score(
feat_index=i, num_neighbors=10),
dtype=int)
idx = idx[np.where(score > self.neighbor_threshold)[0]]
all_ft_idx = np.concatenate((all_ft_idx, idx))
all_ft_idx = np.unique(all_ft_idx)
idxs.append(all_ft_idx)
if len(idxs) == 1:
self.hsic_idx_ = idxs[0]
else:
self.hsic_idx_ = np.intersect1d(idxs[-1], self.hsic_idx_)
print("HSIC done.", len(self.hsic_idx_))
print("Upsampling with ADASYN... (features: " +
str(len(self.hsic_idx_)) + ")")
sm = ADASYN(sampling_strategy="minority",
n_neighbors=self.adasyn_neighbors,
n_jobs=-1)
sX, sy = X[:, self.hsic_idx_], y
if self.adasyn_neighbors > 0:
try:
sX, sy = sm.fit_resample(X[:, self.hsic_idx_], y)
for i in range(len(np.unique(y) - 1)):
sX, sy = sm.fit_resample(sX, sy)
except:
pass
print("ADASYN done. Starting clf")
self.clf_ = LGBMClassifier(n_estimators=1000).fit(sX, sy)
print("done")
return self
class HSICLasso:
def __init__(self, k=10):
self.model = HLasso()
self.k = k
self.modelname = "HSICLasso_{}".format(k)
def fit(self, X, y):
self.model.input(X, y)
self.model.classification(self.k)
self.index = np.array(self.model.get_index())
return self
def transform(self, X):
return X[:, self.index]
def fit_transform(self, X, y):
self.fit(X, y)
return self.transform(X)
def HSICLasso(self):
df_ = self.data.copy()
cols = list(df_.columns)[:-1] + ['class']
df_.columns = cols
hsic_lasso = HSICLasso()
hsic_lasso.input(self.X_train.values, self.Y_train.values)
if self.type == CLASSIFICATION:
hsic_lasso.classification(self.num_top_features)
elif self.type == REGRESSION:
hsic_lasso.regression(self.num_top_features)
feats = [
df_.columns[int(val) - 1] for val in hsic_lasso.get_features()
]
for feat, imp in zip(feats, hsic_lasso.get_index_score()):
features_[feat] = imp
self.report_feature_importance(features_,
self.num_top_features,
label="HSICLasso")
def HSIC_lasso():
hsic = HSICLasso()
hsic.input(data, labels)
before = datetime.datetime.now()
hsic.classification(num_feat=treshold, B=0, M=1, max_neighbors=10, discrete_x=False)
# B a M su na postupne nacitanie ak mam velky dataset, B deli pocet vzoriek, pre klasicky algoritmus B=0, M=1
after = datetime.datetime.now()
print("HSIC Lasso")
selected = hsic.get_index()
print(len(selected))
print("cas: " + str(after - before))
print('\n')
if len(selected) < len(header):
transform_and_save(selected, "HSIC_Lasso")
class ClassificationTest(unittest.TestCase):
def setUp(self):
self.hsic_lasso = HSICLasso()
def test_classification(self):
np.random.seed(0)
with self.assertRaises(UnboundLocalError):
self.hsic_lasso.classification()
self.hsic_lasso.input("test_data/csv_data.csv")
self.hsic_lasso.classification(5, discrete_x=True, n_jobs=1)
self.assertEqual(self.hsic_lasso.A, [764, 1422, 512, 248, 1581])
self.hsic_lasso.input("test_data/csv_data.csv")
self.hsic_lasso.classification(10, discrete_x=True, n_jobs=1)
self.assertEqual(
self.hsic_lasso.A,
[764, 1422, 512, 248, 1581, 1670, 1771, 896, 779, 266])
# Blocks
self.hsic_lasso.input("test_data/csv_data.csv")
B = int(self.hsic_lasso.X_in.shape[1] / 2)
self.hsic_lasso.classification(5, B, 10, discrete_x=True)
self.assertEqual(self.hsic_lasso.A, [764, 1422, 512, 248, 266])
self.hsic_lasso.input("test_data/csv_data.csv")
B = int(self.hsic_lasso.X_in.shape[1] / 2)
self.hsic_lasso.classification(10, B, 10, discrete_x=True)
self.assertEqual(
self.hsic_lasso.A,
[764, 1422, 512, 248, 1670, 1581, 266, 896, 1771, 779])
# use non-divisor as block size
with warnings.catch_warnings(record=True) as w:
self.hsic_lasso.input("test_data/csv_data.csv")
B = int(self.hsic_lasso.X_in.shape[1] / 2) - 1
n = self.hsic_lasso.X_in.shape[1]
numblocks = n / B
self.hsic_lasso.classification(10, B, 10, discrete_x=True)
self.assertEqual(
self.hsic_lasso.A,
[1422, 764, 512, 248, 1670, 1581, 896, 266, 1771, 779])
self.assertEqual(len(w), 1)
self.assertEqual(w[-1].category, RuntimeWarning)
self.assertEqual(
str(w[-1].message), "B {} must be an exact divisor of the \
number of samples {}. Number of blocks {} will be approximated to {}.".format(
B, n, numblocks, int(numblocks)))
def featureSelection(X, y, method = 'lasso', select = 500):
t0 = time.time()
# sparse (15 seconds)
if method == 'lasso':
from sklearn import linear_model
a = 0.861 if select == 500 else 0.0755
lasso = linear_model.Lasso(alpha = a)
lasso.fit(X,y)
XSelected = X[:,lasso.coef_ != 0]
indices = np.where(lasso.coef_ != 0)
if indices > select:
indices = np.argsort(-lasso.coef_)[:select]
# non-sparse (157 seconds)
if method == 'rf':
from sklearn.ensemble import ExtraTreesRegressor
from sklearn.datasets import load_iris
from sklearn.feature_selection import SelectFromModel
t = ExtraTreesRegressor(n_estimators=50)
t.fit(X, y)
model = SelectFromModel(t, prefit=True,
max_features = select)
XSelected = model.transform(X)
indices = np.where(model.get_support)
# non-sparse (8.5 seconds)
if method == 'svm':
from sklearn.svm import SVR
from sklearn.feature_selection import SelectFromModel
SVMReg = SVR(kernel = 'linear',
gamma='scale', C=1.0, epsilon=0.2)
SVMReg.fit(X, y)
model = SelectFromModel(SVMReg, prefit=True,
max_features = select)
XSelected = model.transform(X)
indices = np.where(model.get_support())
# wrapper model (preset number of features) (1000 seconds / 5000 seconds)
if method == 'hsiclasso':
from pyHSICLasso import HSICLasso
hsic_lasso = HSICLasso()
hsic_lasso.input(X,y)
hsic_lasso.regression(select)
XSelected = X[:,hsic_lasso.get_index()]
indices = hsic_lasso.get_index()
# dimensionality reduction
# PCA
# MDS
# PLS
# DWT
# f = h5py.File('selected/' + str(select) + '/X_' + method + '.hdf5', "w")
# f.create_dataset('X', data=XSelected)
# f.create_dataset('indices', data=indices)
# f.close()
# return indices
np.savetxt('selected/' + str(select) + '/X_' + method + '.dat', indices)
# np.savetxt('selected/' + str(select) + '/X_' + method + '.dat', XSelected)
print("--- %s seconds ---" % (time.time() - t0))
def main():
#Numpy array input example
hsic_lasso = HSICLasso()
data = sio.loadmat("../tests/test_data/matlab_data.mat")
X = data['X'].transpose()
Y = data['Y'][0]
featname = ['Feat%d' % x for x in range(1, X.shape[1] + 1)]
hsic_lasso.input(X, Y, featname=featname)
hsic_lasso.regression(5)
hsic_lasso.dump()
hsic_lasso.plot_path()
#Save parameters
hsic_lasso.save_param()
def __init__(self, k=10):
self.model = HLasso()
self.k = k
self.modelname = "HSICLasso_{}".format(k)
def hsic_lasso_matric(self,
data,
n_jobs=2,
n_sample=False,
frac_sample=False):
'''Calculate hsic lasso (subtract correlation between explanatory variables).
Since the correlation coefficient matrix is not symmetric, it is viewed in the row direction.
The correlation between variable 0 and the other variable is stored as the component on the 0th row,
and the correlation between variable 1 and the other variable is stored as the component on the first row.
n_jobs : (int) Indicates the number of cores to be calculated. -1 for GPU.
data: (numpy or pandas) A data frame that contains all explanatory and objective variables
n_sample : (int) How much random sampling to do. False if not.
If a numerical value is entered, sampling is performed using that number of rows.
frac_sample: [0 ~ 1] (float) Sampled as a percentage of the number of rows. Not used at the same time as n_sample.
'''
data = copy(data)
data = pd.DataFrame(data).dropna()
# Sampling when n_sample contains a numerical value
if not n_sample:
if not frac_sample:
# n_sample=False, frac_sample=False
pass
else:
# n_sample=False, frac_sample=int
data = data.sample(frac=frac_sample, replace=True)
else:
if not frac_sample:
# n_sample=int, frac_sample=False
data = data.sample(n=n_sample, replace=True)
else:
# n_sample=int, frac_sample=int
raise ValueError(
'Please enter a value for `frac` OR `n`, not both')
data = check_array(data, accept_sparse="csc",
dtype=float) # Convert to numpy.ndarray
n_col = data.shape[1]
hsic_array = np.empty((0, n_col - 1), float)
for i in range(n_col):
X = np.delete(data, obj=i, axis=1)
y = data[:, i]
# Calculation of hsic_lasso
hsic_lasso = HSICLasso()
hsic_lasso.input(X, y)
hsic_lasso.regression(num_feat=X.shape[1],
discrete_x=False,
n_jobs=n_jobs)
# hsic_lasso only appears in descending order of score, so sort
hsic_ = np.array(
[hsic_lasso.get_index(),
hsic_lasso.get_index_score()])
hsic_ = hsic_.T # Transpose because it is difficult to use
# Since there are not enough scores that came out, add 0.0 to the index to complement
lack_set = set([x for x in range(X.shape[1])]) - set(hsic_[:, 0])
for lack in lack_set:
lack_list = np.array([[lack, 0.0]])
hsic_ = np.append(hsic_, lack_list, axis=0)
hsic_ = hsic_[np.argsort(hsic_[:, 0])] # Sort by index
hsic_array = np.append(hsic_array,
hsic_[:, 1].reshape(1, -1),
axis=0)
# Since it does not include the correlation component with itself, add 1.0
n_row = hsic_array.shape[0]
for i in range(n_row):
insert_i = (n_row + 1) * i
hsic_array = np.insert(hsic_array, insert_i, 1.0)
self.hsic_lasso = hsic_array.reshape(n_row, -1)
return self.hsic_lasso
def main():
hsic_lasso = HSICLasso()
hsic_lasso.input("../tests/test_data/matlab_data.mat")
hsic_lasso.regression(5)
hsic_lasso.dump()
hsic_lasso.plot_path()
class RegressionTest(unittest.TestCase):
def setUp(self):
self.hsic_lasso = HSICLasso()
def test_regression(self):
np.random.seed(0)
with self.assertRaises(UnboundLocalError):
self.hsic_lasso.regression()
self.hsic_lasso.input("./tests/test_data/matlab_data.mat")
self.hsic_lasso.regression(5, n_jobs = 1)
self.assertEqual(self.hsic_lasso.A, [1099, 99, 199, 1299, 299])
self.hsic_lasso.input("./tests/test_data/matlab_data.mat")
self.hsic_lasso.regression(10, n_jobs = 1)
self.assertEqual(self.hsic_lasso.A, [1099, 99, 199, 1299, 1477,
1405, 1073, 299,1596, 358])
# Blocks
self.hsic_lasso.input("./tests/test_data/matlab_data.mat")
B = int(self.hsic_lasso.X_in.shape[1]/2)
self.hsic_lasso.regression(5, B, 10)
self.assertEqual(self.hsic_lasso.A, [1099, 99, 199, 299, 1299])
self.hsic_lasso.input("./tests/test_data/matlab_data.mat")
B = int(self.hsic_lasso.X_in.shape[1]/2)
self.hsic_lasso.regression(10, B, 10)
self.assertEqual(self.hsic_lasso.A, [1099, 99, 199, 1477, 299,
1299, 1073, 1405, 358, 1596])
# use non-divisor as block size
with warnings.catch_warnings(record=True) as w:
self.hsic_lasso.input("./tests/test_data/csv_data.csv")
B = int(self.hsic_lasso.X_in.shape[1]/2) - 1
n = self.hsic_lasso.X_in.shape[1]
numblocks = n / B
self.hsic_lasso.regression(10, B, 10)
self.assertEqual(self.hsic_lasso.A, [1422, 248, 512, 1581, 1670,
764, 1771, 896, 779, 398])
self.assertEqual(len(w), 1)
self.assertEqual(w[-1].category, RuntimeWarning)
self.assertEqual(str(w[-1].message), "B {} must be an exact divisor of the \
number of samples {}. Number of blocks {} will be approximated to {}.".format(B, n, numblocks, int(numblocks)))
#!/usr/bin/env python
import numpy as np
from pyHSICLasso import HSICLasso
hsic_lasso = HSICLasso()
hsic_lasso.input("breast.mat")
hsic_lasso.classification(50)
np.save('features_hl.npy', hsic_lasso.A)
- X_TRAIN: path of a numpy array with x.
- Y_TRAIN: path of a numpy array with y.
- FEATNAMES: path of a numpy array with feature names.
- MODE: regression or classification.
- HL_SELECT: number of features to select.
- HL_B: size of the block.
- HL_M: number of permutations.
Output files:
- features_hl.npy: numpy array with the 0-based index of
the selected features.
'''
import numpy as np
from pyHSICLasso import HSICLasso
hl = HSICLasso()
np.random.seed(0)
hl.X_in = np.load("${X_TRAIN}").T
hl.Y_in = np.load("${Y_TRAIN}").T
hl.Y_in = np.expand_dims(hl.Y_in, 0)
hl.featname = np.load("${FEATNAMES}")
try:
hl.${MODE}($HL_SELECT, B = $HL_B, M = $HL_M, max_neighbors = 50)
except MemoryError:
import sys, traceback
traceback.print_exc()
np.save('features_hl.npy', np.array([]))
sys.exit(77)
def setUp(self):
self.hsic_lasso = HSICLasso()
class InputTest(unittest.TestCase):
def setUp(self):
self.hsic_lasso = HSICLasso()
def test_check_arg(self):
with self.assertRaises(SyntaxError):
self.hsic_lasso._check_args([])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([1, 2, 3])
with self.assertRaises(ValueError):
self.hsic_lasso._check_args(["txt"])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args(["hoge.txt"])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args(["hogecsv"])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([123])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([[1, 2, 3]])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([np.array([1, 2, 3])])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args(["hoge", "hoge"])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args(["hoge", [1, 2, 3]])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([[1, 2, 3], "hoge"])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args(["hoge", np.array([1, 2, 3])])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([np.array([1, 2, 3]), "hoge"])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([123, [1, 2, 3]])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([[1, 2, 3], 123])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([123, np.array([1, 2, 3])])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([np.array([1, 2, 3]), 123])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([[1, 2, 3], np.array([1, 2, 3])])
with self.assertRaises(TypeError):
self.hsic_lasso._check_args([np.array([1, 2, 3]), [1, 2, 3]])
self.assertTrue(self.hsic_lasso._check_args(["hoge.csv"]))
self.assertTrue(self.hsic_lasso._check_args(["hoge.tsv"]))
self.assertTrue(self.hsic_lasso._check_args(["hoge.mat"]))
self.assertTrue(
self.hsic_lasso._check_args(
[np.array([1, 2, 3]), np.array([1, 2, 3])]))
self.assertTrue(self.hsic_lasso._check_args([[1, 2, 3], [1, 2, 3]]))
def test_input_data_file(self):
self.assertTrue("./tests/test_data/csv_data.csv")
self.assertTrue("./tests/test_data/tsv_data.tsv")
self.assertTrue("./tests/test_data/mat_data.mat")
def test_input_data_list(self):
self.hsic_lasso._input_data_list([[1, 1, 1], [2, 2, 2]], [1, 2])
X_in_row, X_in_col = self.hsic_lasso.X_in.shape
Y_in_row, Y_in_col = self.hsic_lasso.Y_in.shape
self.assertEqual(X_in_row, 3)
self.assertEqual(X_in_col, 2)
self.assertEqual(Y_in_row, 1)
self.assertEqual(Y_in_col, 2)
self.hsic_lasso._input_data_list([[1, 1, 1, 1, 1], [2, 2, 2, 2, 2],
[3, 3, 3, 3, 3], [4, 4, 4, 4, 4]],
[1, 2, 3, 4])
X_in_row, X_in_col = self.hsic_lasso.X_in.shape
Y_in_row, Y_in_col = self.hsic_lasso.Y_in.shape
self.assertEqual(X_in_row, 5)
self.assertEqual(X_in_col, 4)
self.assertEqual(Y_in_row, 1)
self.assertEqual(Y_in_col, 4)
with self.assertRaises(ValueError):
self.hsic_lasso._input_data_list(
[[1, 1, 1, 1, 1], [2, 2, 2, 2, 2], [3, 3, 3, 3, 3],
[4, 4, 4, 4, 4]], [[1, 2, 3, 4], [1, 2, 3, 4]])
def test_input_data_ndarray(self):
self.hsic_lasso._input_data_ndarray(np.array([[1, 1, 1], [2, 2, 2]]),
np.array([1, 2]))
X_in_row, X_in_col = self.hsic_lasso.X_in.shape
Y_in_row, Y_in_col = self.hsic_lasso.Y_in.shape
self.assertEqual(X_in_row, 3)
self.assertEqual(X_in_col, 2)
self.assertEqual(Y_in_row, 1)
self.assertEqual(Y_in_col, 2)
self.hsic_lasso._input_data_ndarray(
np.array([[1, 1, 1, 1, 1], [2, 2, 2, 2, 2], [3, 3, 3, 3, 3],
[4, 4, 4, 4, 4]]), np.array([1, 2, 3, 4]))
X_in_row, X_in_col = self.hsic_lasso.X_in.shape
Y_in_row, Y_in_col = self.hsic_lasso.Y_in.shape
self.assertEqual(X_in_row, 5)
self.assertEqual(X_in_col, 4)
self.assertEqual(Y_in_row, 1)
self.assertEqual(Y_in_col, 4)
with self.assertRaises(ValueError):
self.hsic_lasso._input_data_list(
np.array([[1, 1, 1, 1, 1], [2, 2, 2, 2, 2], [3, 3, 3, 3, 3],
[4, 4, 4, 4, 4]]),
np.array([[1, 2, 3, 4], [1, 2, 3, 4]]))
def test_check_shape(self):
self.hsic_lasso._input_data_list([[1, 1, 1], [2, 2, 2]], [1, 2])
self.assertTrue(self.hsic_lasso._check_shape())
self.hsic_lasso._input_data_list([[1, 1, 1, 1, 1], [2, 2, 2, 2, 2],
[3, 3, 3, 3, 3], [4, 4, 4, 4, 4]],
[1, 2, 3, 4])
self.assertTrue(self.hsic_lasso._check_shape())
self.hsic_lasso._input_data_list(
[[1, 1, 1, 1], [2, 2, 2, 2], [3, 3, 3, 3], [4, 4, 4, 4]],
[1, 2, 3, 4])
self.assertTrue(self.hsic_lasso._check_shape())
self.hsic_lasso._input_data_list(
[[1, 1, 1, 1, 1], [2, 2, 2, 2, 2], [3, 3, 3, 3, 3]], [1, 2, 3, 4])
with self.assertRaises(ValueError):
self.hsic_lasso._check_shape()
def test_input(self):
self.assertTrue(
self.hsic_lasso.input("./tests/test_data/csv_data.csv"))
self.assertTrue(
self.hsic_lasso.input("./tests/test_data/tsv_data.tsv"))
self.assertTrue(
self.hsic_lasso.input("./tests/test_data/matlab_data.mat"))
def hsic(train, test, K):
hsic_lasso = HSICLasso()
hsic_lasso.input(train[0], train[1])
hsic_lasso.classification(K, n_jobs=-1)
indices = hsic_lasso.get_index()
return indices
def fit(self, X, y):
if X.shape[1] > 10000:
#clf = RandomForestClassifier(n_estimators=1000,n_jobs=-1).fit(X,y)
clf = LGBMClassifier(n_estimators=1000, n_jobs=-1).fit(X, y)
ftimp = clf.feature_importances_
relevant = np.where(ftimp > 0)[0]
print("relevant ft:", len(relevant), "/", X.shape[1])
else:
relevant = np.arange(X.shape[1])
sss = StratifiedShuffleSplit(n_splits=self.hsic_splits,
random_state=42)
idxs = []
hsics = []
for train_index, test_index in list(sss.split(X, y)):
hsic_lasso2 = HSICLasso()
hsic_lasso2.input(X[:, relevant][train_index], y[train_index])
hsic_lasso2.classification(
self.n_features, B=self.B,
M=self.M) #(self.n_features, B=self.B, M=self.M)
hsics.append(hsic_lasso2)
# not just best features - get their neighbors (similar features) too
all_ft_idx = np.array(hsic_lasso2.get_index(), dtype=int).ravel()
for i in range(len(all_ft_idx)):
idx = np.array(hsic_lasso2.get_index_neighbors(
feat_index=i, num_neighbors=10),
dtype=int)
score = np.array(hsic_lasso2.get_index_neighbors_score(
feat_index=i, num_neighbors=10),
dtype=int)
idx = idx[np.where(score > self.neighbor_threshold)[0]]
all_ft_idx = np.concatenate((all_ft_idx, idx))
all_ft_idx = np.unique(all_ft_idx)
idxs.append(relevant[all_ft_idx])
#if len(idxs) == 1:
# self.hsic_idx_ = idxs[0]
#else:
# self.hsic_idx_ = np.intersect1d(idxs[-1], self.hsic_idx_)
self.hsic_idx_ = []
stability_concession = 0
while len(self.hsic_idx_) == 0:
featurecandidates = np.unique(np.concatenate(idxs))
for candidate in featurecandidates:
occurrences = np.sum(
[1 if candidate in idx else 0 for idx in idxs])
if occurrences > self.stability_minimum_across_splits - stability_concession:
self.hsic_idx_.append(candidate)
if len(self.hsic_idx_) > 1:
break
else:
# failed to find commonly occurring features - reduce threshold
stability_concession += 1
print("HSIC done.", len(self.hsic_idx_), "(out of ",
len(featurecandidates), " candidates)")
print("Upsampling with ADASYN... (features: " +
str(len(self.hsic_idx_)) + ")")
sm = ADASYN(sampling_strategy="minority",
n_neighbors=self.adasyn_neighbors,
n_jobs=-1)
sX, sy = X[:, self.hsic_idx_], y
if self.adasyn_neighbors > 0:
try:
sX, sy = sm.fit_resample(X[:, self.hsic_idx_], y)
for i in range(len(np.unique(y) - 1)):
sX, sy = sm.fit_resample(sX, sy)
except:
pass
print("ADASYN done. Starting clf")
self.clf_ = LGBMClassifier(n_estimators=1000).fit(sX, sy)
print("done")
return self
def main():
hsic_lasso = HSICLasso()
#out_list = ['c'+str(i) for i in range(1,51)]
#print (out_list)
hsic_lasso.input("./user_data_new.csv",
output_list=[
'c1', 'c2', 'c3', 'c4', 'c5,', 'c6', 'c7', 'c8', 'c9',
'c10'
])
# ,'c11', 'c12', 'c13', 'c14', 'c15,', 'c16', 'c17', 'c18', 'c19', 'c20','c21', 'c22', 'c23', 'c24', 'c25,', 'c26', 'c27', 'c28', 'c29', 'c30'])
hsic_lasso.regression(100, B=50)
hsic_lasso.dump()
select_index = hsic_lasso.get_index()
print(select_index)
print(hsic_lasso.get_index_score())
#hsic_lasso.plot_path()
print(hsic_lasso.get_features())
X_select = hsic_lasso.X_in[select_index, :]
np.savetxt('X_select.txt', X_select, fmt=str('%.5f'), encoding='utf-8')
from pyHSICLasso import HSICLasso
hsic_lasso = HSICLasso()
hsic_lasso.input("SNR-26415.csv")
print(hsic_lasso.classification(100))
hsic_lasso.get_features()
l = []
l.append(hsic_lasso.get_features())
print(hsic_lasso.get_features())
print(len(l))
temp = 0
hsic_lasso.dump()
for i in range(0, len(l)):
print(l[i])
temp = temp + 1
print(temp)
- Y_TRAIN: path of a numpy array with y.
- COVARS: path of a numpy array with covariates.
- FEATNAMES: path of a numpy array with feature names.
- MODE: regression or classification.
- HL_SELECT: number of features to select.
- HL_B: size of the block.
- HL_M: number of permutations.
Output files:
- features_hl.npy: numpy array with the 0-based index of
the selected features.
'''
import numpy as np
from pyHSICLasso import HSICLasso
hl = HSICLasso()
np.random.seed(0)
hl.X_in = np.load("${X_TRAIN}").T
hl.Y_in = np.load("${Y_TRAIN}").T
hl.Y_in = np.expand_dims(hl.Y_in, 0)
hl.featname = np.load("${FEATNAMES}")
covars = np.load('${COVARS}')
try:
hl.${MODE}($HL_SELECT, B = $HL_B, M = $HL_M, covars = covars)
except MemoryError:
import sys, traceback
traceback.print_exc()
np.save('features_hl.npy', np.array([]))
sys.exit(77)