for paradigm, scores in split_paradigm_score(
df, args.folder, path, category):
statistics = fault_statistics(scores, statistics,
category,
name + paradigm)
else:
statistics = fault_statistics(df, statistics, category,
name)
if not statistics.empty:
save_dataframe(statistics, folder, category, False)
def fault_statistics(df, statistics, category, name):
print(f'[{category}] Fault-statistics: {name}')
total_rows = len(df)
faulty_rows = len(df[df['faults'] > 0])
non_faulty_rows = len(df[df['faults'] == 0])
percentage_faulty = (faulty_rows / total_rows) * 100
result = {
'name': name,
'rows': total_rows,
'faulty_rows': faulty_rows,
'non_faulty_rows': non_faulty_rows,
'percentage_faulty': percentage_faulty
}
return statistics.append(result, ignore_index=True)
if __name__ == '__main__':
main(parse_args())
from analysis import descriptive, fault_statistics, univariate, multivariate, parse_args, multivariate_baseline, \
multivariate_baseline_hasdata, multivariate_baseline_control, fault_metric_statistics
if __name__ == '__main__':
args = parse_args()
descriptive.main(args)
fault_statistics.main(args)
fault_metric_statistics.main(args)
univariate.main(args)
multivariate.main(args)
if args.multivariate_baseline:
multivariate_baseline.main(args)
multivariate_baseline_hasdata.main(args)
if args.multivariate_baseline_control:
multivariate_baseline_control.main(args)