forked from sophiaalthammer/dossier_coliee
-
Notifications
You must be signed in to change notification settings - Fork 0
/
nano.save
13 lines (13 loc) · 812 Bytes
/
nano.save
1
2
3
4
5
6
7
8
9
10
11
12
13
#!/bin/bashcreate_index(){
arg1=$1
python -m pyserini.index -collection JsonCollection -generator DefaultLuceneDocumentGenerator -threads 1 -input /home/salthamm/data/coliee19/task1/task1_train/$arg1/ -index /home/salthamm/data/coliee19/task1/task1_train/$arg1/index/ -storePositions -storeDocvectors -storeRaw
}
for sub_dir in {200..285}
do
echo $sub_dir
create_index $sub_dir
done
python coliee_task1_search_index.py --train-dir /home/salthamm/data/coliee19/task1/task1_train/
# here the searching fails for too long indices - store these indices with a try and except and then try with shorter size again
python coliee_task1_eval_index.py --train-dir /home/salthamm/data/coliee19/task1/task1_train/
# reaches recall of 0.9322899536843366 on coliee task 1 train set