コード例 #1
0
print "reduced_feature_matrix_logistic before transpose:%d" % len(reduced_feature_matrix_logistic)	
reduced_feature_matrix_logistic = zip(*reduced_feature_matrix_logistic)
print "reduced_feature_matrix_logistic after transpose:%d" % len(reduced_feature_matrix_logistic)

coef_path_linear_cv.fit(reduced_feature_matrix_logistic,y)

linear_cv_score = cross_validation.cross_val_score(coef_path_linear_cv, reduced_feature_matrix_logistic, y, n_jobs=2, cv=CV, scoring=Scoring)

linear_results_parameters = [ coef_path_linear_cv.predict(reduced_feature_matrix_logistic), coef_path_linear_cv.get_params,reduced_feature_list_logistic, coef_path_linear_cv.coef_]

linear_scores = [linear_cv_score, r2_score(y, linear_results_parameters[0]), 'linear']

print "reduced_feature_list_logistic length:%d" % len(reduced_feature_list_logistic)
print "linear_coefficient length:%d" % len(coef_path_linear_cv.coef_)
linear_word_results = []
###LINEAR REGRESSION END

for i in range(len(coef_path_linear_cv.coef_)):
	temp_list = [reduced_feature_list_logistic[i], coef_path_linear_cv.coef_[i]]
	linear_word_results.append(temp_list)

#word_priority_linear = sorted (linear_word_results, key= lambda x: float(x[1]), reverse=True)

model_results = [forest_results_parameters, lasso_results_parameters, elastic_results_parameters, logistic_results_parameters, binary_x_logistic_results_parameters, linear_results_parameters]

model_scores = [forest_scores, lasso_scores, elastic_scores, logistic_scores, binary_x_logistic_scores, linear_scores]

post_processing(model_results, model_scores, X, y, widget_selection, list_of_features, Ngram_Range_Low, Ngram_Range_High)

コード例 #2
0
reduced_feature_matrix_logistic = zip(*reduced_feature_matrix_logistic)
print "reduced_feature_matrix_logistic after transpose:%d" % len(reduced_feature_matrix_logistic)

coef_path_linear_cv.fit(reduced_feature_matrix_logistic,y)

linear_cv_score = cross_validation.cross_val_score(coef_path_linear_cv, reduced_feature_matrix_logistic, y, n_jobs=2, cv=CV, scoring=Scoring)

linear_results_parameters = [ coef_path_linear_cv.predict(reduced_feature_matrix_logistic), coef_path_linear_cv.get_params,reduced_feature_list_logistic, coef_path_linear_cv.coef_]

linear_scores = [linear_cv_score, r2_score(y, linear_results_parameters[0]), 'linear']

print "reduced_feature_list_logistic length:%d" % len(reduced_feature_list_logistic)
print "linear_coefficient length:%d" % len(coef_path_linear_cv.coef_)
linear_word_results = []
###LINEAR REGRESSION END

for i in range(len(coef_path_linear_cv.coef_)):
	temp_list = [reduced_feature_list_logistic[i], coef_path_linear_cv.coef_[i]]
	linear_word_results.append(temp_list)

#model_results = [forest_results_parameters, lasso_results_parameters, elastic_results_parameters, logistic_results_parameters, 0, linear_results_parameters]

#model_scores = [forest_scores, lasso_scores, elastic_scores, logistic_scores, 0, linear_scores]
model_results = [forest_results_parameters, 0, 0, logistic_results_parameters, 0, linear_results_parameters]

model_scores = [forest_scores, 0, 0, logistic_scores, 0, linear_scores]

post_processing(model_results, model_scores, X, y, widget_selection, list_of_features, 
		Ngram_Range_Low, Ngram_Range_High, Min_DF, PageLoaded, WidgetViewed, ite)