def set_kpca_obj(pars, data, sntype, type_number):
"""
Set kpca object based on cross-validation results.
input: pars, dict
output from read_hyperpar
data, array
output from read_matrix
sntype, list
output from read_matrix
type_number, dict
dictionary to translate types between raw data and final
classification
keywords -> final classificaton elements
values -> identifiers in raw data
output: obj_kpca, KernelPCA obj
tailored with cross-validation results
spec_matrix, array
low dimension spectroscopic matrix
labels, list
classes as defined in raw data files
"""
from sklearn.decomposition import KernelPCA
import numpy as np
# start kpca object
obj_kpca = KernelPCA()
obj_kpca.eigen_solver = pars['eigen_solver']
obj_kpca.kernel = pars['kernel']
obj_kpca.alpha = pars['alpha']
obj_kpca.gamma = pars['gamma']
obj_kpca.n_components = pars['n_components']
obj_kpca.coef0 = pars['coef0']
obj_kpca.degree = pars['degree']
obj_kpca.tol = pars['tol']
obj_kpca.fit_inverse_transform = pars['fit_inverse_transform']
obj_kpca.remove_zero_eig = pars['remove_zero_eig']
obj_kpca.kernel_params = pars['kernel_params']
obj_kpca.max_iter = pars['max_iter']
spec_matrix = obj_kpca.fit_transform(data)
# construct label vector
labels = []
for elem in sntype:
for classes in type_number.keys():
if elem in type_number[classes]:
labels.append(classes)
labels = np.array(labels)
return obj_kpca, spec_matrix, labels
def compute_scores(X):
pca = PCA(svd_solver='auto')
kpca = KernelPCA(fit_inverse_transform=True)
ica = FastICA()
nmf = NMF(init='nndsvda')
pca_scores, ica_scores, nmf_scores, kpca_scores = [], [], [], []
for n in n_components:
pca.n_components = n
ica.n_components = n
nmf.n_components = n
kpca.n_components = n
print(n)
Xpca = pca.inverse_transform(pca.fit_transform(Xs))
pca_scores.append(explained_variance_score(Xs, Xpca))
Xica = ica.inverse_transform(ica.fit_transform(Xs))
ica_scores.append(explained_variance_score(Xs, Xica))
Xkpca = kpca.inverse_transform(kpca.fit_transform(Xs))
kpca_scores.append(explained_variance_score(Xs, Xkpca))
Xnmf = nmf.inverse_transform(nmf.fit_transform(X))
nmf_scores.append(explained_variance_score(X, Xnmf))
return pca_scores, ica_scores, nmf_scores, kpca_scores
