From c4d98ead445a4c874345256cb4b71752b2ff9c91 Mon Sep 17 00:00:00 2001 From: lintool Date: Fri, 6 Dec 2024 07:40:12 -0500 Subject: [PATCH] Removed 'NoMerge' statement in docs for BEIR. --- ...ons-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ssions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...ressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...eir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...ns-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...essions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...gressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...r-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...s-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...ns-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ....0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...r-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...stack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...upstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...qadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...r-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...stack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...upstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...qadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ....0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ....0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...ir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...pstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...dupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...eir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...adupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ....0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ....0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...adupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ....0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.md | 2 -- ....0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...k-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...ack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...pstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...dupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...r-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...stack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...upstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...qadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...adupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ....0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.md | 2 -- ....0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...k-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...ack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...pstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...dupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...r-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...eir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...upstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...adupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ....0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...eir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...adupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ....0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ....0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ....0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...dupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...qadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...qadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...ck-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...tack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...upstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...adupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ....0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...ack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...stack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...dupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...qadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...r-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...s-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...bpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...ons-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...sions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...egressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...ir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...ns-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...ions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ssions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...gressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...r-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...eir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...s-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...ons-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ns-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...sions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...essions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ....0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...ir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ns-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...sions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...essions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ....0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...ir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...ssions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.md | 2 -- .../regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...eir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...ons-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...sions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...ons-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...sions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...egressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...ir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...ns-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ns-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...sions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...essions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ....0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...ir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...s-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ons-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ssions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...ressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...eir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...s-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ons-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ssions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...ressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...eir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ns-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...sions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...essions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ....0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...ir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...eir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ons-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...sions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ....0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ....0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...ir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...s-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...ssions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...r-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...eir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ....0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...ir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...is-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.md | 2 -- ...ebis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md | 2 -- ...0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.md | 2 -- ...0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.md | 2 -- ...ressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...egressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- .../regressions-dl19-passage.bge-base-en-v1.5.hnsw.cached.md | 2 -- .../regressions-dl19-passage.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md | 2 -- ...ions-dl19-passage.cohere-embed-english-v3.0.hnsw.cached.md | 2 -- ...egressions-dl19-passage.cos-dpr-distil.hnsw-int8.cached.md | 2 -- .../regressions-dl19-passage.cos-dpr-distil.hnsw-int8.onnx.md | 2 -- .../regressions-dl19-passage.cos-dpr-distil.hnsw.cached.md | 2 -- .../regressions-dl19-passage.cos-dpr-distil.hnsw.onnx.md | 2 -- .../regressions-dl19-passage.openai-ada2.hnsw-int8.cached.md | 2 -- .../regressions-dl19-passage.openai-ada2.hnsw.cached.md | 2 -- ...ressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...egressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- .../regressions-dl20-passage.bge-base-en-v1.5.hnsw.cached.md | 2 -- .../regressions-dl20-passage.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md | 2 -- ...ions-dl20-passage.cohere-embed-english-v3.0.hnsw.cached.md | 2 -- ...egressions-dl20-passage.cos-dpr-distil.hnsw-int8.cached.md | 2 -- .../regressions-dl20-passage.cos-dpr-distil.hnsw-int8.onnx.md | 2 -- .../regressions-dl20-passage.cos-dpr-distil.hnsw.cached.md | 2 -- .../regressions-dl20-passage.cos-dpr-distil.hnsw.onnx.md | 2 -- .../regressions-dl20-passage.openai-ada2.hnsw-int8.cached.md | 2 -- .../regressions-dl20-passage.openai-ada2.hnsw.cached.md | 2 -- ...ns-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.md | 2 -- ...ions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.md | 2 -- ...essions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.md | 2 -- ...gressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.md | 2 -- ...o-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md | 2 -- ...smarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.md | 2 -- ...ions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.md | 4 +--- ...ssions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.md | 2 -- ...gressions-msmarco-v1-passage.cos-dpr-distil.hnsw.cached.md | 4 +--- ...regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.md | 2 -- ...essions-msmarco-v1-passage.openai-ada2.hnsw-int8.cached.md | 2 -- .../regressions-msmarco-v1-passage.openai-ada2.hnsw.cached.md | 2 -- ...ir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ....0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...r-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...eir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ....0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...r-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...e-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...limate-fever.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...upstack-android.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...adupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.template | 2 -- ....0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...android.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...k-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...tack-android.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...pstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...upstack-english.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...adupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.template | 2 -- ....0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...english.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...k-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...tack-english.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...pstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...dupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...qadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ck-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...stack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...upstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ....0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...ack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...stack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...dupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...qadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...ack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...stack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...dupstack-mathematica.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...qadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...ematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...thematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...ck-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...upstack-physics.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...adupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.template | 2 -- ....0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...k-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...tack-physics.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...pstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...ack-programmers.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...stack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...dupstack-programmers.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...qadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...rammers.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ogrammers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...-programmers.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...ck-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...adupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ....0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.template | 2 -- ....0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...k-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...pstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...dupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ....0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...ack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...stack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...dupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...qadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...qadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...ck-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...tack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...upstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...adupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...tack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...pstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...adupstack-webmasters.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...masters.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ebmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...k-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...ack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...stack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...upstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...qadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...rdpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...ck-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...tack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ....0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...pedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...ir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...r-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...eir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- .../beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ....0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ....0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...ir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...r-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...otpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...r-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...fcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- .../beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- .../beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ....0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...ir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...ir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...r-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...r-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...obust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...ir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ....0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...ir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ....0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...r-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...ignal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ....0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...ir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...c-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...rec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...eir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...ec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ....0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ....0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- ...ebis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- ...0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.template | 2 -- ...1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...che2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.template | 2 -- ...ouche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template | 2 -- ...s-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.template | 2 -- ...bis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.template | 2 -- .../dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- .../dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../dl19-passage.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../dl19-passage.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...assage.cohere-embed-english-v3.0.hnsw-int8.cached.template | 2 -- ...l19-passage.cohere-embed-english-v3.0.hnsw.cached.template | 2 -- .../dl19-passage.cos-dpr-distil.hnsw-int8.cached.template | 2 -- .../dl19-passage.cos-dpr-distil.hnsw-int8.onnx.template | 2 -- .../dl19-passage.cos-dpr-distil.hnsw.cached.template | 2 -- .../templates/dl19-passage.cos-dpr-distil.hnsw.onnx.template | 2 -- .../dl19-passage.openai-ada2.hnsw-int8.cached.template | 2 -- .../templates/dl19-passage.openai-ada2.hnsw.cached.template | 2 -- .../dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- .../dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../dl20-passage.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../dl20-passage.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...assage.cohere-embed-english-v3.0.hnsw-int8.cached.template | 2 -- ...l20-passage.cohere-embed-english-v3.0.hnsw.cached.template | 2 -- .../dl20-passage.cos-dpr-distil.hnsw-int8.cached.template | 2 -- .../dl20-passage.cos-dpr-distil.hnsw-int8.onnx.template | 2 -- .../dl20-passage.cos-dpr-distil.hnsw.cached.template | 2 -- .../templates/dl20-passage.cos-dpr-distil.hnsw.onnx.template | 2 -- .../dl20-passage.openai-ada2.hnsw-int8.cached.template | 2 -- .../templates/dl20-passage.openai-ada2.hnsw.cached.template | 2 -- ...arco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.template | 2 -- ...smarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.template | 2 -- .../msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.template | 2 -- .../msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.template | 2 -- ...assage.cohere-embed-english-v3.0.hnsw-int8.cached.template | 2 -- ...-v1-passage.cohere-embed-english-v3.0.hnsw.cached.template | 2 -- ...smarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.template | 4 +--- .../msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.template | 2 -- .../msmarco-v1-passage.cos-dpr-distil.hnsw.cached.template | 4 +--- .../msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.template | 2 -- .../msmarco-v1-passage.openai-ada2.hnsw-int8.cached.template | 2 -- .../msmarco-v1-passage.openai-ada2.hnsw.cached.template | 2 -- 534 files changed, 4 insertions(+), 1072 deletions(-) diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.md index b9f7ef8bc..c26232b6a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.md index 829fb3f94..366c2feaf 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.md index e3c5fc421..5a7d6636b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 824afae17..1fae42c76 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 17e376afa..d0b2c6bcb 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.md index 5ca0dd68f..80ea1be2b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.md index cf377630b..87ce56132 100644 --- a/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-arguana.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.md index f192011d7..35a824c8d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.md index 52dbaf72d..4db9d8c9b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.md index 0c1e906aa..6fadc3991 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.md index 8d0d7900f..5c949eba9 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 43f2014f6..96050ac18 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index bd9e1f3af..bac170a5b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.md index 70025ec42..9991d0bab 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.md index 76821c481..9755ffbc7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-bioasq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.md index 99ba07bbf..6b5efdb8b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.md index a5683611d..b816052c8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.md index ebc27e474..91de9ec30 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.md index a0fbc75be..2789378ab 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index bdc0e8fe5..1093c18d3 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 77ca2640a..02bca6b29 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.md index 3bbd75682..02951f5c6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md index 3bec20c39..513dc05ec 100644 --- a/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-climate-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.md index 06231676f..04f5a8529 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.md index 79c1ce8f5..13f906e8f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.md index 5eec210e4..c0e710f9c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.md index d3ec442f8..4d80b6cfb 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index d32c98c88..6ff2b3e8a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index e93a5db47..0ee7ba662 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.md index 1610a8f3e..af41f5a1f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.md index 320c8c9bc..d52fa515f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.md index 7510eea05..4b1bd1132 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.md index a3a16be19..80d48ccb8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.md index 66a023dd1..97fe77187 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.md index 29df371aa..97b1e4dac 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index e5992e48e..da4963fc4 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 0ab069fa3..32c5d7234 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.md index 3c3c2a15f..7be510c72 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.md index e35847ce2..003e1e58f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.md index aaf82ca54..47c9aada8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.md index 8d5b36018..3f8d3e2e5 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.md index 7795ea1cd..a324869b7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.md index 62898bef1..cdf241422 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 4efb81b63..970d26803 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index dcd2f1096..473022bc9 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.md index d59a3ec30..e49c9aca4 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.md index d00c8659c..fc57da76d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.md index a2536c328..5dcb28adc 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.md index 4627561f6..b055202d4 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.md index ae28f3301..9c1f917e6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.md index f20a17ae8..8e371e532 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index c112d4667..fb2834349 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 8bab1db22..21d746f70 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.md index 1a8ea33a7..3e66de74c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.md index 47e2b252d..38f44c399 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.md index 2637de72c..bd644db23 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.md index b1428529f..289bab59b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.md index e80cd577b..dbc1decf4 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.md index 806eceaf6..b323a7db5 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index f7b1978bd..75913a569 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 203d35b6c..d68531393 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.md index 7c88f8c80..cf47a4ba4 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.md index 6ae1bd115..1574e17c8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.md index 52533ae47..1a8c637d2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.md index bc920d0f2..b81e06747 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.md index a9711a8c8..6055837f7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.md index 4dbfc032e..2e8677662 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index dbd13863f..4ab0d763a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 1f54bf8a0..dd877ae08 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.md index d633d983a..5429f0c1e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.md index 645222db9..cca8d53db 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.md index 1274727c2..964d46380 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.md index 82e41fc16..4f4ce675b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.md index 0ca720161..8429dbc86 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.md index 5aaf8ff4a..065aa415e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 273832b4c..84bb09445 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index c59ff7179..e04d32a82 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.md index c7001d0f5..e34555ed3 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.md index ae0b42d7b..6a57a45b6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.md index 3cfd2e242..275473490 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.md index e533a9acd..2e1c498e2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.md index b8a2e543a..6b7287bdf 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.md index 52e03ee83..a5a55f408 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index ac204f47b..910e5b946 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 016db3302..a6370d962 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.md index c9b61e8d2..b2f24d416 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.md index c951c35a9..684f50a9e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.md index 0523d7293..1fbd76273 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.md index 7bed88635..18caa44d8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.md index 65fe36632..2266bf0a2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.md index a5ee08ccc..3c7235669 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 542737177..3a392f47a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index bf5a9e373..025885a74 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.md index 0cd1fe7f7..f9c84ccb9 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.md index 886dd28c4..f07c42abf 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.md index 817c0823e..ce0bd732c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.md index 534a04b00..37149190f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.md index 80100178f..0a64e76ed 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.md index f23c9e55e..705e2b78d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 9ef1745a8..48a73567a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 43d5198ff..99ecb62fa 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.md index 41e169ce9..4b0d20fc8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.md index 8d15a42f1..cdb1e8d9a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.md index 6af6704ee..038547a3a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.md index 30938cfcc..e5877c355 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.md index 89af998db..b15820de2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.md index f071b50e2..590c4429f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 135338142..2d76d5791 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index c642228d4..0edb3c81d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.md index 9ab22eeff..6578d0ee8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.md index 873f875a0..4d9238071 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.md index d1ada9394..a764b0c19 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.md index 1c163e7de..80e759db1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.md index f9abcf58f..ef01c51af 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.md index 3dda3fa69..c17564a1f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 20da877a5..40d80eee5 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 2001a1fc1..62eef2823 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.md index ab31974e9..ab647aaee 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.md index 5ca1b18a1..b1db26655 100644 --- a/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.md index 304fb7edf..65f1009a2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.md index 3ebfae18a..7c9ed9813 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.md index ad891c1dd..5746ae3d2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.md index 6362b85b1..4d899c50d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index ed4dfe935..bd9b811eb 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 2bacd5175..01e704078 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.md index eb3640660..f180f6c8f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.md index 2b11b6bed..6dabf7528 100644 --- a/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.md index dfa0ac23e..e63f22e63 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.md index 759719300..bf4a123dd 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.md index f8c2d2690..dc81c0a83 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.md index 2279bae55..190e190aa 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index cde6a1bde..c14abb576 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 000103759..973e2b65a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.md index 8f185d8ff..4d6a575af 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md index 5cf0a02ab..6b9aa3e41 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fever.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.md index 3ea4104aa..faa8923a9 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.md index 6d4a6a84a..a208a8e11 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.md index 0013ad3a6..6f81eb8a1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.md index 474ad0165..de7d3171e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 03a8fbde1..862ac46f1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 03037cefc..220f4205a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.md index 6a184b35e..980cb8a9c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.md index 16c3b06a3..d985a579f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-fiqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.md index b52a36037..25c92af4d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.md index df44fc900..4df6febfd 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.md index c6718351e..72fd508d7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.md index f663fa52e..b0910de63 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 7e6fcc574..74af354e7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index a3810e8e4..f6469c201 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.md index e201a4ef1..a040af689 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.md index 3b180e5f6..c04bff9c1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-hotpotqa.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.md index e78bd328b..c9c24d68b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.md index 225c3c53e..88dbe98ac 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.md index 4d8cad2f8..c1ba9f0b6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.md index 1079cb410..3549379d1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 2005d7fb7..ef9189977 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 9913f9f6f..a525e7058 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.md index 9afad2c11..3e051555c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.md index 0b480fc79..04896521a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nfcorpus.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.md index 2c7757fe9..94e90dc2c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.md index 468964082..45c1639f6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.md index 606f2ce9b..83889ab6c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.md index dd3727d88..0519b0410 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 3205974e1..3c56d7b85 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 1434b80c9..e870953fb 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.md index d88868b6f..8ac8fbe3b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.md index e1fe23e02..36fc7634a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-nq.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.md index a490f67cb..d88d9985c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.md index 7c4681a60..5b82f8eb3 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.md index 45b69122d..ea87718fd 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.md index c45af9a78..085efc4ad 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index d90a75596..f3058feb1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index caf07b221..98f0a5df6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.md index 1381c83f6..8bc626a63 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.md index 4618aafe0..9d6007c23 100644 --- a/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-quora.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.md index 53dd3b2a9..a33eca954 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.md index 4e7c2aac5..6c630ba98 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.md index 03ebe7a0c..d2b51ec62 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.md index 5db8ee453..db1f27ce7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 03e9068d7..29ca82be6 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 68e9b560c..a521181c4 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.md index c0591f10d..254b5b80d 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.md index 98e30d6e6..8b8aae7d8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-robust04.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.md index f2ef14b67..407a59ecd 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.md index 5d8f4dfc3..51bb595f3 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.md index b120f5b87..15eecf7d5 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.md index 7b83b5e1a..e10fe9d43 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 1cb3dd499..a5b3e038f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index a021f0566..39483df28 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.md index 21bfeb2e3..4da7be97b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.md index 9830e6a87..00fd9c569 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scidocs.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.md index 02d22e47a..3d8ace4f2 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.md index 0ce045398..a286e666c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.md index a554e924b..38cbcc3a1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.md index 5f7995e5d..3e7a91619 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 0b245f928..697342d0a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index f2c5b62f9..e3184ca82 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.md index f80a440e0..de6f8c514 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.md index 51034c957..f37017f07 100644 --- a/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-scifact.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.md index 21b596b5c..eeaebff02 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.md index 895ddc50c..d2acd68e1 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.md index cae3fed06..54bdeba29 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.md index 643c1803b..8275e85b0 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 9ecbece3c..dfb965f54 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 8fa0a3043..179edf696 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.md index 9ec1b9e6d..2b799fde5 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.md index daffa2ed4..11c50e6e8 100644 --- a/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-signal1m.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.md index 17bc9a24b..911c34f82 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.md index f04b7a59c..ca1e51bce 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.md index 2a6c88e50..62e320c29 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.md index 3670bfdf3..2e742aa7a 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 3969d6a75..322b736d3 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index 90ff92c0b..66da8660c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.md index fab0ddb37..17962280b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.md index f491f064c..b9ed8a1f9 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-covid.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.md index 36582e283..ea33b6074 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.md index 6f0bfc2f6..5abfef24f 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.md index 4405b5d2c..26f3bb3c7 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.md index a387afdfd..a40e5398b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index 5f07bdafb..7096779f9 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index ac988b277..e04a25a1e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.md index 6ad519c3c..794f63d3b 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.md index 2690ff20e..22c966936 100644 --- a/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-trec-news.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.md index 7e3275b73..46c22cb85 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.md index b94d99447..017dc77bc 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.md index 28aa2c178..c62e06245 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.md index e7e09717b..0391ede7e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.md index a373df051..36e34476c 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md index e3af27baa..081e93773 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.md index dae72b2e9..f08a1f0be 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.md b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.md index 69d065b22..a5b45136e 100644 --- a/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.md +++ b/docs/regressions/regressions-beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.md @@ -43,8 +43,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ ``` The path `/path/to/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.md index 13c9c2645..f8932ee61 100644 --- a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.md index e464cadbf..dfe6b2ec9 100644 --- a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.cached.md index d45864761..895d4d7c5 100644 --- a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.onnx.md index 3e4ce7de4..103a8f057 100644 --- a/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-dl19-passage.bge-base-en-v1.5.hnsw.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md b/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md index 150e5aca5..31100672d 100644 --- a/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md @@ -60,8 +60,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cohere-embed-english-v3.0/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw.cached.md b/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw.cached.md index c536817ca..1d3329f89 100644 --- a/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw.cached.md +++ b/docs/regressions/regressions-dl19-passage.cohere-embed-english-v3.0.hnsw.cached.md @@ -60,8 +60,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cohere-embed-english-v3.0/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.cached.md b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.cached.md index 82a67e248..8c52a3568 100644 --- a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.onnx.md b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.onnx.md index ba80901e3..2d3c1a32d 100644 --- a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw-int8.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.cached.md b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.cached.md index 4d80d8ea8..ab1fbc829 100644 --- a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.cached.md +++ b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.onnx.md b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.onnx.md index dd0b79bad..9db00f428 100644 --- a/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.onnx.md +++ b/docs/regressions/regressions-dl19-passage.cos-dpr-distil.hnsw.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw-int8.cached.md b/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw-int8.cached.md index 2da259d06..7e2a9d13c 100644 --- a/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw-int8.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-openai-ada2/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw.cached.md b/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw.cached.md index 15f3b32f7..6e1373668 100644 --- a/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw.cached.md +++ b/docs/regressions/regressions-dl19-passage.openai-ada2.hnsw.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-openai-ada2/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.md index db28c5e83..2ae1b6159 100644 --- a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.md index b5fabdaf2..1a41467d6 100644 --- a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.cached.md index bd37bd12c..e4c98a18b 100644 --- a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.onnx.md index 4d5252c21..f449c76ba 100644 --- a/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-dl20-passage.bge-base-en-v1.5.hnsw.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md b/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md index 477d55b2d..7343b9ad2 100644 --- a/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md @@ -60,8 +60,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cohere-embed-english-v3.0/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw.cached.md b/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw.cached.md index 0d4bcbec2..305157492 100644 --- a/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw.cached.md +++ b/docs/regressions/regressions-dl20-passage.cohere-embed-english-v3.0.hnsw.cached.md @@ -60,8 +60,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cohere-embed-english-v3.0/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.cached.md b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.cached.md index 33aded5f3..74deab191 100644 --- a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.onnx.md b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.onnx.md index cd06e2401..2b5444244 100644 --- a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw-int8.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.cached.md b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.cached.md index 803c73de0..2da624aef 100644 --- a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.cached.md +++ b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.onnx.md b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.onnx.md index 9dda2d476..96eea55ea 100644 --- a/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.onnx.md +++ b/docs/regressions/regressions-dl20-passage.cos-dpr-distil.hnsw.onnx.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw-int8.cached.md b/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw-int8.cached.md index 17136e5c6..3ee98985b 100644 --- a/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw-int8.cached.md +++ b/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw-int8.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-openai-ada2/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw.cached.md b/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw.cached.md index c7aeff6a4..4257f91bd 100644 --- a/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw.cached.md +++ b/docs/regressions/regressions-dl20-passage.openai-ada2.hnsw.cached.md @@ -65,8 +65,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-openai-ada2/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.md b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.md index 221f82e61..f91f6d50c 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.md b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.md index a2a59c221..3b85d708f 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.md b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.md index aa659b7e3..7936bda14 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.md b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.md index 62a43cda7..72b26fe42 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.md +++ b/docs/regressions/regressions-msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-bge-base-en-v1.5/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md b/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md index ec0059a8f..a9caa19e0 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.md @@ -60,8 +60,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cohere-embed-english-v3.0/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.md b/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.md index 0a2537a0a..4b8b6e470 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.md @@ -60,8 +60,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cohere-embed-english-v3.0/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.md b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.md index 2c55e93ee..b1f08502e 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. @@ -114,4 +112,4 @@ Note that both HNSW indexing and quantization are non-deterministic (i.e., resul To add to this reproduction log, modify [this template](../../src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.template) and run `bin/build.sh` to rebuild the documentation. -+ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) \ No newline at end of file ++ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) diff --git a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.md b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.md index a63ab27a2..215b23ae0 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.md +++ b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.cached.md b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.cached.md index 26d84c79e..e50869083 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.cached.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval @@ -112,4 +110,4 @@ Note that HNSW indexing is non-deterministic (i.e., results may differ slightly To add to this reproduction log, modify [this template](../../src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.cached.template) and run `bin/build.sh` to rebuild the documentation. -+ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) \ No newline at end of file ++ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) diff --git a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.md b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.md index e34438c6e..dab73d2c7 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.md +++ b/docs/regressions/regressions-msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-cos-dpr-distil/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw-int8.cached.md b/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw-int8.cached.md index 5535884e4..ecf2959c9 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw-int8.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw-int8.cached.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-openai-ada2/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw.cached.md b/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw.cached.md index 24a40d1b1..aec65c441 100644 --- a/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw.cached.md +++ b/docs/regressions/regressions-msmarco-v1-passage.openai-ada2.hnsw.cached.md @@ -62,8 +62,6 @@ bin/run.sh io.anserini.index.IndexHnswDenseVectors \ The path `/path/to/msmarco-passage-openai-ada2/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.template index 7ba229ebf..4d5093e65 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.template index 1830438ba..595049265 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.template index 2c9d9fbb0..ecfd79c90 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 12f6a86a8..645c509bd 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 69dd22fff..259bcb332 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.template index 2580b6696..beedc8f33 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.template index 4dbe4820e..7bf7cea38 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-arguana.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.template index 05c7a3eac..3b4d1cd99 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.template index 140cba972..33d53fd21 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.template index 50bb40272..f21b09c35 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.template index 2fa38bf7c..e5183bb6d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index c8becd906..15e327a21 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index b893840cf..bdbade175 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.template index b92fe4489..fb2216507 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.template index 1b21d4585..fb30025a0 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-bioasq.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.template index a171d3e6c..cd4fc22d4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.template index 4715ee4c1..9ccb241df 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.template index 75173695a..f428c0dce 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.template index edab40c98..024ed53f4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 5242596b7..daac9477e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index b071a91b5..e1e00f135 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.template index e2351fd5d..5009a98d5 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template index 31b1e7b16..f87b2c119 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-climate-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.template index 1b06763a7..ace60211b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.template index af12ee23a..a244da255 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.template index 720d20afe..9786b9ee8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.template index 150b8c2ee..3c318a439 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 29bba2c01..951d9cdb2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 22fbecee7..06ae00df1 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.template index dca85fdf6..1797dfae2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.template index 084226d12..d8e07d2b4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.template index 8c42137f2..bfebdfe81 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.template index 62cd23ca6..b0414190b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.template index bee4b6481..ecab07e23 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.template index b607b8cf3..003991070 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 0167601b0..e4ebf6005 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index a96f1fdc2..9fba48479 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.template index 32d849c8a..e865597e6 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.template index 633e1169b..9676f5eb5 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.template index e96dec3f6..743d44c87 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.template index 81c621052..08da168c8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.template index 311fac23c..8ce05a97d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.template index f456d74e0..6f6e18dd1 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index a87d8720a..9d8b65300 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 9f41f50da..c323ee47b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.template index aa222cb34..c35cef625 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.template index 697ab2551..177081b9b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.template index a0126c1da..56d74f03a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.template index d02b9cfe3..19a3fa527 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.template index c4b2b7991..d85f06989 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.template index 95893dfb6..b179874d2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 4c7029ebc..31c8e554e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index af1aa345a..c61f95042 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.template index 316c84dda..b9890c788 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.template index bdd8082ea..ef991980e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.template index 9e29a7c65..4e9cdc3f7 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.template index 07f968bc6..9f5a265b4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.template index 0154151db..4621e6d7b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.template index a48f68c53..05546c03f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 64a747323..7d8942f68 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 856add95e..54d707c9b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.template index 24afdc730..d140ca1fa 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.template index 4e24e2e4d..2e362b4fd 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.template index bab4374f3..fd44c354c 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.template index e65420d91..d6251a792 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.template index 8b370b536..b7b94923a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.template index 95bfb6c2b..b5e94b897 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 805f321c5..8326f6795 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 8307ac9c1..88d7bee73 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.template index eee371275..c1c7f681e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.template index ad4468b1a..12bd1e186 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.template index 095b3ada0..fe0ee1a19 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.template index 665d3dd55..cb9509b85 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.template index c068b90fc..7e0a12ab4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.template index 84ad5d419..9f3041dae 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index c52814944..a9bc61a64 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 2ac2447c8..5edcfdf02 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.template index db81559fc..f4905f212 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.template index 7f99cbed9..59a7271f2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.template index 635ebd6fb..1bd1f70bc 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.template index a78a77654..499c3ca57 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.template index 348f10c27..95f7253d3 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.template index 831c42641..d59869a82 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 37291b7cd..af210cba1 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index e5766809b..ab86a0a0a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.template index 6626fbb72..8cdbffb8a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.template index 6c534c0ac..f1c9e4f94 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.template index 5da2d26ee..afab2ecea 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.template index 0e83c6fd5..736837cca 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.template index 5664170fc..17857be5d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.template index b6d6a1e23..0dd3d20d0 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index ee423f7b2..dfa63e7b6 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 9c582f6b9..dc7b9dbbe 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.template index ce01d2342..4fecd54c2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.template index 4a3169a61..b9c0470ea 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.template index e2030dd20..e14d8423b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.template index 9c3d9edee..650e58c07 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.template index 69d7293b9..ae2eccdec 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.template index af3ca19ae..13d054792 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index af4998954..99f2cc4e5 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 5b95735dc..af93cbf7d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.template index d5aef804d..b5e336de4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.template index 17393bc57..0eeacd7f9 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.template index af1d7a750..43e6edca4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.template index 49cf301a7..bf54495ad 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.template index 91e2d7887..20dd53fc6 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.template index f69422801..2ff059005 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 7abf6fd7a..0cc4dd043 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index f4a89049e..63a804abd 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.template index cee9ccb39..fd9fedd77 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.template index 6196fb1d7..8f4dcfe0b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.template index a9683d37d..cbc2e9069 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.template index 744e087d4..8b6d1b01a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.template index 17c022670..aab26c15a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.template index a68486b20..a401f66a2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index d1bfb602a..a5820d0fc 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index a0498eda4..73fc9cdc0 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.template index fd13e23bb..3ba64adc0 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.template index a54a28e62..a52d05b74 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.template index ceaf01f48..95a7b9e02 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.template index e53caf82c..124be044d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.template index d07261324..59c7d4dd2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.template index 806536bce..ee3e35665 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 940a769af..cc04373cb 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 354984040..aa23f877a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.template index 6edaab631..f83025111 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.template index 0b90c0f0e..2b16bb094 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.template index f8d682a85..98add75c8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.template index ab6b3a854..1e550c4ab 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.template index 8669ad648..dfec8240a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.template index 34e47e9ae..1b945e9bc 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 81126b308..b5ce08007 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index a74af8c81..bcae36b24 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.template index 21014f2a8..c24c311f2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template index b6d5ddc77..089d31656 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fever.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.template index 2d364c347..cca5962a8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.template index 581fe36ed..dc63df611 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.template index 40a2a8c0c..2770a872f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.template index ffa3dc23a..76435bd6f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 2c433d3d3..ccea6b6ac 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index a76b7d1ed..be4a12b1d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.template index 94aeed5a6..59f5c4967 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.template index ba952d22d..c3caa2cf7 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-fiqa.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.template index 5dafd5080..6aeae33d8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.template index 6f7a38b77..1f80e46b7 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.template index 0d37ffeae..83ee8e85f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.template index 030237464..5d50d544b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 10e6b9d7a..d1b670c6a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 958e4d0d6..510e983c3 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.template index dfd8773d5..46c2f2d51 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.template index 18fdf1fa8..eb8b4d4b2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-hotpotqa.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.template index e6e8ee16c..b8cd992dd 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.template index fac764fee..f63a7368e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.template index 833a5523f..f7b1b1604 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.template index e1b27d207..baf120016 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index f3150bf1c..435c67cdf 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 4e23a97f6..e47580389 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.template index 95554889d..b78757473 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.template index 68d755f06..83e22e798 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nfcorpus.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.template index 5f9e43179..ce95dfb3c 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.template index 7b34ccc5b..740adb833 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.template index 2cba96579..d2eb5b35a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.template index d3fd44561..21f0cdfb3 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 56c67e782..0606002ea 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 91fed1950..0bf47f423 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.template index 874ca1b81..47b65f8da 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.template index 5ae86b04a..9d1629125 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-nq.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.template index 882347841..68d99e751 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.template index ea3ea0986..7968efa1c 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.template index f17061238..11d421f61 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.template index 5dd2c02b2..10f07d422 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 5366975aa..57c806bc3 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 902f4bee2..576485782 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.template index ca82e1db7..61cc10a34 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.template index ffbec1438..e4d64fc75 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-quora.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.template index c32ce1e27..3f7ca0fe1 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.template index 689621d8d..66b5bdc49 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.template index 0676d5273..457db7391 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.template index df04002b2..e7ee94b7d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 0d69a84b0..4049fb57c 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 46977479c..ced60e4cb 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.template index cc9182019..3f5b3d1d0 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.template index 86f95778a..4ecf87166 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-robust04.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.template index 4c2af7107..177fe8dba 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.template index a2a1b49c4..73af136dc 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.template index bf00680cc..a8fae5040 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.template index bd7a47363..4007dce6e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 44b20d191..0d9e83406 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 65f1ad814..9bc9c0c7a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.template index 348a1569b..1bb3b2089 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.template index e1ae055fb..d587deef7 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scidocs.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.template index 7b07cf11c..674ada9ac 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.template index 4ed7f33e8..a68076cec 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.template index 749e4f32c..0425f30d1 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.template index 07fb4416a..bb577cb33 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index b634fd2cf..173efaa2e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 808af590c..804c1c66e 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.template index 5a990396b..a213a5b88 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.template index 635748427..754eef240 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-scifact.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.template index 704e22cd0..05f4ad040 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.template index 83ce303ac..a879dc266 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.template index 171fff269..c42060f6f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.template index db40b335d..d427a8d8f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 06e956671..de2b62400 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 10c332c3e..6b77a614a 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.template index 79bf2dc69..c7d3168a7 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.template index 92a8ef06a..79d9cfbdf 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-signal1m.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.template index 0c4f35b33..7c7fc5030 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.template index d9b95458a..485132a42 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.template index 1ab8be5c6..ee401d968 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.template index 5429c2300..99de33aaf 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 3275b84d8..7df878aae 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index b4a6fe2b3..81fe7af76 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.template index cc60affe4..21ddb01d8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.template index 04da38e36..70a02a485 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-covid.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.template index bad558d3d..a8556838c 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.template index 37841d21a..30a0c344b 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.template index bd364bc5b..b79bdab2d 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.template index 3353032f2..4a515cf98 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 953b8128e..48d9b44e3 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 69b156314..c5a25f024 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.template index aef306add..9fc4f2548 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.template index 224e45183..56a392ae4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-trec-news.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.template index 379c52183..bd0b1b9e4 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.template index f5851d20e..3a9e470f0 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.template index 90641553f..b5852adf5 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.template index 081620596..3d1acbf8f 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.template index 80714d118..61979f341 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template index 0cb3b786c..14ada0df8 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw-int8.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.template index fc597d529..796aecaee 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.cached.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.template b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.template index d516f9b62..cf36e67f2 100644 --- a/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/beir-v1.0.0-webis-touche2020.bge-base-en-v1.5.parquet.hnsw.onnx.template @@ -36,8 +36,6 @@ ${index_cmds} ``` The path `/path/to/${corpus}/` should point to the corpus downloaded above. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.template index 63fdad31d..11d7f8a4c 100644 --- a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.template index 69afbfc03..dcd7a050b 100644 --- a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.cached.template index d011773dd..d258062b1 100644 --- a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.onnx.template index 89c9d8ee7..296f1c195 100644 --- a/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/dl19-passage.bge-base-en-v1.5.hnsw.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template index 565c13fea..5ed06e455 100644 --- a/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template @@ -53,8 +53,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw.cached.template b/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw.cached.template index a47174b5f..5bfb8df92 100644 --- a/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.cohere-embed-english-v3.0.hnsw.cached.template @@ -53,8 +53,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.cached.template index 9406fe382..c55d96074 100644 --- a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.onnx.template index ff348cdbc..d9429c26e 100644 --- a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw-int8.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.cached.template b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.cached.template index e387bf030..41405cdc7 100644 --- a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.onnx.template b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.onnx.template index c6d7c0e52..d9a0c9b14 100644 --- a/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/dl19-passage.cos-dpr-distil.hnsw.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw-int8.cached.template index 991983ff4..045df030c 100644 --- a/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw-int8.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw.cached.template b/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw.cached.template index 24ec176d9..50add8dc0 100644 --- a/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl19-passage.openai-ada2.hnsw.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.template index a67e15082..956790300 100644 --- a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.template index 957322a5b..b197088a4 100644 --- a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.cached.template index 9197b856b..18fdd1ebb 100644 --- a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.onnx.template index 058502495..6a81774ac 100644 --- a/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/dl20-passage.bge-base-en-v1.5.hnsw.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template index fcc92a8dc..d65eba96b 100644 --- a/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template @@ -53,8 +53,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw.cached.template b/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw.cached.template index 1bc8727a0..9b40bcf5a 100644 --- a/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.cohere-embed-english-v3.0.hnsw.cached.template @@ -53,8 +53,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.cached.template index 4c9517f5c..2bfaf45fa 100644 --- a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.onnx.template index 28dc4082c..0cd21c0e6 100644 --- a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw-int8.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.cached.template b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.cached.template index fd89cd5bd..95e75285a 100644 --- a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.onnx.template b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.onnx.template index 582cc4609..fc0256d56 100644 --- a/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/dl20-passage.cos-dpr-distil.hnsw.onnx.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw-int8.cached.template b/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw-int8.cached.template index ad2259f77..7687266f8 100644 --- a/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw-int8.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw.cached.template b/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw.cached.template index 361c8bb69..c2d930581 100644 --- a/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw.cached.template +++ b/src/main/resources/docgen/templates/dl20-passage.openai-ada2.hnsw.cached.template @@ -58,8 +58,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.template index 486b4bc69..4758ea506 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.cached.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.template index 5dc942633..2052635a6 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw-int8.onnx.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.template index aba6e3262..8689ff2e0 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.cached.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.template b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.template index fb6a69722..b6818458b 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.bge-base-en-v1.5.hnsw.onnx.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template index 924906b79..18b91d7fa 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw-int8.cached.template @@ -53,8 +53,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.template index 5f51caffa..0799aa04b 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.cohere-embed-english-v3.0.hnsw.cached.template @@ -53,8 +53,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.template index 89099d71f..5a915df83 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.cached.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. @@ -91,4 +89,4 @@ Note that both HNSW indexing and quantization are non-deterministic (i.e., resul To add to this reproduction log, modify [this template](${template}) and run `bin/build.sh` to rebuild the documentation. -+ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) \ No newline at end of file ++ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.template b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.template index b3737d545..ed45e4ec3 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw-int8.onnx.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.cached.template index 5173b9b09..635b1f86a 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.cached.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval @@ -89,4 +87,4 @@ Note that HNSW indexing is non-deterministic (i.e., results may differ slightly To add to this reproduction log, modify [this template](${template}) and run `bin/build.sh` to rebuild the documentation. -+ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) \ No newline at end of file ++ Results reproduced by [@yilinjz](https://github.com/yilinjz) on 2023-09-01 (commit [`4ae518b`](https://github.com/castorini/anserini/commit/4ae518bb284ebcba0b273a473bc8774735cb7d19)) diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.template b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.template index 3defd04a8..6f5f60a2d 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.cos-dpr-distil.hnsw.onnx.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw-int8.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw-int8.cached.template index b82567bab..eb88ab10c 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw-int8.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw-int8.cached.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. Furthermore, we are using Lucene's [Automatic Byte Quantization](https://www.elastic.co/search-labs/blog/articles/scalar-quantization-in-lucene) feature, which increase the on-disk footprint of the indexes since we're storing both the int8 quantized vectors and the float32 vectors, but only the int8 quantized vectors need to be loaded into memory. See [issue #2292](https://github.com/castorini/anserini/issues/2292) for some experiments reporting the performance impact. diff --git a/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw.cached.template b/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw.cached.template index a1827f765..6d7a43a57 100644 --- a/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw.cached.template +++ b/src/main/resources/docgen/templates/msmarco-v1-passage.openai-ada2.hnsw.cached.template @@ -55,8 +55,6 @@ ${index_cmds} The path `/path/to/${corpus}/` should point to the corpus downloaded above. Upon completion, we should have an index with 8,841,823 documents. -Note that here we are explicitly using Lucene's `NoMergePolicy` merge policy, which suppresses any merging of index segments. -This is because merging index segments is a costly operation and not worthwhile given our query set. ## Retrieval