X = scal.fit_transform(X)
X, y = utils.shuffle(X, y, random_state=5)
###############
method_name = 'GPR-set'
# method_name = 'SVR-set'
# method_name = 'KNR-set'
# method_name = 'KRR-set'
# method_name = 'PAR-set'
# method_name = 'SGDR-set'
# method_name = 'BRR-set'
# method_name = 'LASSO-set'
# method_name = 'AdaBR-em'
# method_name = 'GBR-em'
method_name = 'DTR-em'
me1, cv1, scoring1, param_grid1 = method = dict_method_reg()[method_name]
estimator = GridSearchCV(me1,
cv=cv1,
scoring=scoring1,
param_grid=param_grid1,
n_jobs=1)
n_select = (2, 3)
clf = Exhaustion(estimator,
n_select=n_select,
muti_grade=2,
muti_index=[2, X.shape[1]],
must_index=None,
n_jobs=4,
refit=True).fit(X, y)
X_frame['destiny'] = x_rame
"""base_method"""
# method_name = ['GPR-set', 'SVR-set', 'KRR-set']
method_name = [
'GPR-set', 'SVR-set', 'KRR-set', 'KNR-set', 'GBR-em', 'AdaBR-em',
'RFR-em', "DTR-em"
]
index_all = [
data.pickle_pd.GPR_set23, data.pickle_pd.SVR_set23,
data.pickle_pd.KRR_set23
]
estimator_all = []
for i in method_name:
me1, cv1, scoring1, param_grid1 = dict_method_reg()[i]
estimator_all.append(
GridSearchCV(me1,
cv=cv1,
scoring=scoring1,
param_grid=param_grid1,
n_jobs=1))
"""union"""
# [print(_[0]) for _ in index_all]
index_slice = [
tuple(index[0]) for _ in index_all for index in _[:round(len(_) / 3)]
]
index_slice = list(set(index_slice))
index_slice = [
index_slice[i] for i in [100, 8, 6, 152, 81, 106, 129, 19, 73, 170]
]