def _modeler(train_inds, test_inds, flux_arr, model, method):
new_model = est_clone(model)
flux_train = flux_arr[train_inds]
#print("Training new model: " + str(new_model))
if method == 'KPCA':
new_model.fit_inverse_transform = False
new_model.fit(flux_train)
sqrt_lambdas = np.diag(np.sqrt(model.lambdas_))
X_transformed = np.dot(model.alphas_, sqrt_lambdas)
n_samples = X_transformed.shape[0]
K = new_model._get_kernel(X_transformed)
K.flat[::n_samples + 1] += model.alpha
new_model.dual_coef_ = linalg.solve(K, flux_train)
new_model.X_transformed_fit_ = X_transformed
new_model.fit_inverse_transform = True
else:
if np.any(np.isinf(flux_train)):
print("The given flux_train array contains inf's!")
if np.any(np.isnan(flux_train)):
print("The given flux_train array contains nan's!")
new_model.fit(flux_train)
#print("Returning new model: " + str(new_model))
return new_model
def _iter_modeler(train_inds, test_inds, flux_arr, model, method):
model_list = []
flux_train = flux_arr[train_inds]
flux_avg = np.mean(flux_train, axis=0)
for i in range(2):
new_model = est_clone(model)
new_model.fit(flux_train)
back_train = iz.transform_inverse_transform(flux_train, new_model, flux_avg, method)
flux_train -= back_train
model_list.append(new_model)
return model_list, flux_avg
