def cv(theme, percentage, current_svm):
	[dataset, features] = parse_theme(theme)
	[known_dataset, known_targets, unk] = split_dataset(dataset, targets)
	known_targets = np.asarray(known_targets)

	# cv_features = features_cross_validation(known_dataset, known_targets, features, current_svm)
	# selected_features = select_final_features_from_cv(cv_features, percentage)
	selected_features = select_features(percentage, theme)

	sf = SelectedFeatures(known_dataset, known_targets, selected_features, features)
	combined_dataset = sf.extract_data_from_selected_features()

	std = StandardizedData(known_targets, combined_dataset)
	known_dataset_scaled, known_targets = std.split_and_standardize_dataset()  

	print '####### FEATURES ####### %d \n %s' % (len(selected_features), str(selected_features)) 	
	return cross_validation(np.array(known_dataset_scaled), known_targets, ids, current_svm)
Пример #2
0
        return np.ma.masked_array(np.interp(value, x, y))

if __name__ == "__main__":
	spreadsheet = Spreadsheet(project_data_file)
	data = Data(spreadsheet)
	targets = data.targets
	ids = data.ids

	theme = raw_input("Theme.\n")
	percentage = float(raw_input("Percentage as float.\n"))

	[dataset, features] = parse_theme(theme)
	[known_dataset, known_targets, unk] = split_dataset(dataset, targets)
	known_targets = np.asarray(known_targets)
	
	selected_features = select_features(percentage, theme)
	sf = SelectedFeatures(known_dataset, known_targets, selected_features, features)
	dataset = sf.extract_data_from_selected_features()

	dataset = preprocessing.scale(dataset)

	C_range = np.arange(0.1, 9, 0.1)
	gamma_range = np.arange(0.1, 9, 0.1)
	param_grid = dict(gamma=gamma_range, C=C_range)
	# cv = StratifiedShuffleSplit(known_targets, random_state=42)
	cv = StratifiedKFold(known_targets, n_folds=10)
	grid = GridSearchCV(SVC(class_weight='auto'), param_grid=param_grid, cv=cv, scoring='f1')
	grid.fit(dataset, known_targets)
	print("The best parameters are %s with a score of %0.2f" % (grid.best_params_, grid.best_score_))

	classifiers = []