Use train subset by default in scripts.ndcg

scripts/common.py

@@ -7,7 +7,7 @@ import numpy as np
 from .configuration import EVALUATORS, PARSED_RELEVANCE_JUDGEMENTS
 
 
-def ndcg(parsed_run, task='task1', subset='test'):
+def ndcg(parsed_run, task='task1', subset='train'):
     only_judged_parsed_run = deepcopy(parsed_run)
     parsed_relevance_judgements = PARSED_RELEVANCE_JUDGEMENTS[subset][task]
     for topic_name, results in parsed_run.items():

scripts/evaluate.py

@@ -21,7 +21,7 @@ if __name__ == '__main__':
                 result_name = re.sub('_', ', ', os.path.basename(result)[:-4])
                 with open(result, 'rt') as f:
                     parsed_result = parse_run(f)
-                user_results.append((ndcg(parsed_result, task), result_name))
+                user_results.append((ndcg(parsed_result, task, 'test'), result_name))
             best_ndcg, best_result_name = max(user_results)
             task_results.append((best_ndcg, user_name, best_result_name))
             with open(os.path.join(user, 'README.md'), 'wt') as f: