Tweak regression settings (castorini#2038)

+ Minor tweaks to order in which we're running regressions + Make use of tjena
dannychn11 · Dec 19, 2022 · 65983be · 65983be
1 parent abc14ab
commit 65983be
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 17 deletions.
diff --git a/src/main/python/regressions-batch04.txt b/src/main/python/regressions-batch04.txt
@@ -3,7 +3,14 @@ python src/main/python/run_regression.py --index --verify --search --regression
 python src/main/python/run_regression.py --index --verify --search --regression car17v2.0 > logs/log.car17v2.0 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression car17v2.0-doc2query > logs/log.car17v2.0-doc2query 2>&1
 
-# HC4
+# Newswire
+python src/main/python/run_regression.py --index --verify --search --regression disk12 > logs/log.disk12 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression disk45 > logs/log.disk45 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression robust05 > logs/log.robust05 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression core17 > logs/log.core17 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression core18 > logs/log.core18 2>&1
+
+# HC4 (part 1)
 python src/main/python/run_regression.py --index --verify --search --regression hc4-v1.0-fa > logs/log.hc4-v1.0-fa 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression hc4-v1.0-ru > logs/log.hc4-v1.0-ru 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression hc4-v1.0-zh > logs/log.hc4-v1.0-zh 2>&1
@@ -16,29 +23,13 @@ python src/main/python/run_regression.py --index --verify --search --regression
 python src/main/python/run_regression.py --index --verify --search --regression hc4-neuclir22-ru-en > logs/log.hc4-neuclir22-ru-en 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression hc4-neuclir22-zh-en > logs/log.hc4-neuclir22-zh-en 2>&1
 
-python src/main/python/run_regression.py --index --verify --search --regression neuclir22-fa-qt > logs/log.neuclir22-fa-qt 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression neuclir22-fa-dt > logs/log.neuclir22-fa-dt 2>&1
-
-python src/main/python/run_regression.py --index --verify --search --regression neuclir22-ru-qt > logs/log.neuclir22-ru-qt 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression neuclir22-ru-dt > logs/log.neuclir22-ru-dt 2>&1
-
-python src/main/python/run_regression.py --index --verify --search --regression neuclir22-zh-qt > logs/log.neuclir22-zh-qt 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression neuclir22-zh-dt > logs/log.neuclir22-zh-dt 2>&1
-
 # Other
 python src/main/python/run_regression.py --index --verify --search --regression fever > logs/log.fever 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression wikipedia-dpr-100w-bm25 > logs/log.wikipedia-dpr-100w-bm25 2>&1
 
 python src/main/python/run_regression.py --index --verify --search --regression mb11 > logs/log.mb11 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression mb13 > logs/log.mb13 2>&1
 
-# Newswire
-python src/main/python/run_regression.py --index --verify --search --regression disk12 > logs/log.disk12 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression disk45 > logs/log.disk45 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression robust05 > logs/log.robust05 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression core17 > logs/log.core17 2>&1
-python src/main/python/run_regression.py --index --verify --search --regression core18 > logs/log.core18 2>&1
-
 # Other CLIR
 python src/main/python/run_regression.py --index --verify --search --regression ntcir8-zh > logs/log.ntcir8-zh 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression clef06-fr > logs/log.clef06-fr 2>&1
@@ -47,6 +38,18 @@ python src/main/python/run_regression.py --index --verify --search --regression
 python src/main/python/run_regression.py --index --verify --search --regression fire12-hi > logs/log.fire12-hi 2>&1
 python src/main/python/run_regression.py --index --verify --search --regression fire12-en > logs/log.fire12-en 2>&1
 
+# HC4 (part 2)
+# e.g., hc4-neuclir22-ru-en and neuclir22-ru-dt both write to indexes/lucene-index.neuclir22-ru-en so we need to stagger
+
+python src/main/python/run_regression.py --index --verify --search --regression neuclir22-fa-qt > logs/log.neuclir22-fa-qt 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression neuclir22-fa-dt > logs/log.neuclir22-fa-dt 2>&1
+
+python src/main/python/run_regression.py --index --verify --search --regression neuclir22-ru-qt > logs/log.neuclir22-ru-qt 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression neuclir22-ru-dt > logs/log.neuclir22-ru-dt 2>&1
+
+python src/main/python/run_regression.py --index --verify --search --regression neuclir22-zh-qt > logs/log.neuclir22-zh-qt 2>&1
+python src/main/python/run_regression.py --index --verify --search --regression neuclir22-zh-dt > logs/log.neuclir22-zh-dt 2>&1
+
 # Background linking
 python src/main/python/run_regression.py --verify --search --regression backgroundlinking18 > logs/log.backgroundlinking18 2>&1
 python src/main/python/run_regression.py --verify --search --regression backgroundlinking19 > logs/log.backgroundlinking19 2>&1

diff --git a/src/main/python/run_regression.py b/src/main/python/run_regression.py
@@ -46,6 +46,7 @@
 # There is no need to specify them on a per-file basis.
 CORPUS_ROOTS = [
     '',                           # here, stored in this directory
+    '/mnt/',                      # on tjena
     '/tuna1/',                    # on tuna
     '/store/',                    # on orca
     '/collection',                # on hops (1st location)