def create_features_labels_df_test():
cr = svcomp15.read_category('static/results-xml-raw', 'mixed-examples')
sourcefiles = {i for k in cr.keys() for i in cr[k].index}
features_df = f.create_feature_df(sourcefiles)
# RPC dataset
ranking_df = ranking.create_ranking_df(cr, svcomp15.compare_results)
features_ranking_df = f.create_features_labels_df(features_df, ranking_df)
# Learning via Utility Functions datasets
score_dfdict = classification.create_benchmark_score_dfdict(cr, svcomp15.score)
features_score_dfdict = {b: f.create_features_labels_df(features_df,
score_dfdict[b]) for b in score_dfdict.keys()}
cputime_dfdict = regression.create_benchmark_cputime_dfdict(cr)
features_cputime_dfdict = {b: f.create_features_labels_df(features_df,
cputime_dfdict[b]) for b in cputime_dfdict.keys()}
def extract_ranking_df(xml_dir, category, max_size):
results = svcomp15.read_category(xml_dir, category, max_size)
return ranking.create_ranking_df(results, svcomp15.compare_results)
def create_benchmark_ranking_df_test():
category_results = svcomp15.read_category('static/results-xml-raw', 'mixed-examples')
df = ranking.create_ranking_df(category_results, svcomp15.compare_results)
df.to_csv('ranking_df.csv')
