BEIR

Key:

BM25 Flat BM25 Multifield SPLADE Contriever MSMARCO BGE-base
nDCG@10 R@100 nDCG@10 R@100 nDCG@10 R@100 nDCG@10 R@100 nDCG@10 R@100
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-trec-covid.flat \
  --topics beir-v1.0.0-trec-covid-test \
  --output run.beir.bm25-flat.trec-covid.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-covid-test \
  run.beir.bm25-flat.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-covid-test \
  run.beir.bm25-flat.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-covid-test \
  run.beir.bm25-flat.trec-covid.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-trec-covid.multifield \
  --topics beir-v1.0.0-trec-covid-test \
  --output run.beir.bm25-multifield.trec-covid.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-covid-test \
  run.beir.bm25-multifield.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-covid-test \
  run.beir.bm25-multifield.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-covid-test \
  run.beir.bm25-multifield.trec-covid.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-trec-covid.splade-pp-ed \
  --topics beir-v1.0.0-trec-covid.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.trec-covid.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-covid-test \
  run.beir.splade-pp-ed.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-covid-test \
  run.beir.splade-pp-ed.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-covid-test \
  run.beir.splade-pp-ed.trec-covid.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-trec-covid.contriever-msmarco \
  --topics beir-v1.0.0-trec-covid-test \
  --output run.beir.contriever-msmarco.trec-covid.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-covid-test \
  run.beir.contriever-msmarco.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-covid-test \
  run.beir.contriever-msmarco.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-covid-test \
  run.beir.contriever-msmarco.trec-covid.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-trec-covid.bge-base-en-v1.5 \
  --topics beir-v1.0.0-trec-covid-test \
  --output run.beir.bge-base-en-v1.5.trec-covid.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-covid-test \
  run.beir.bge-base-en-v1.5.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-covid-test \
  run.beir.bge-base-en-v1.5.trec-covid.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-covid-test \
  run.beir.bge-base-en-v1.5.trec-covid.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-bioasq.flat \
  --topics beir-v1.0.0-bioasq-test \
  --output run.beir.bm25-flat.bioasq.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-bioasq-test \
  run.beir.bm25-flat.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-bioasq-test \
  run.beir.bm25-flat.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-bioasq-test \
  run.beir.bm25-flat.bioasq.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-bioasq.multifield \
  --topics beir-v1.0.0-bioasq-test \
  --output run.beir.bm25-multifield.bioasq.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-bioasq-test \
  run.beir.bm25-multifield.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-bioasq-test \
  run.beir.bm25-multifield.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-bioasq-test \
  run.beir.bm25-multifield.bioasq.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-bioasq.splade-pp-ed \
  --topics beir-v1.0.0-bioasq.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.bioasq.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-bioasq-test \
  run.beir.splade-pp-ed.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-bioasq-test \
  run.beir.splade-pp-ed.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-bioasq-test \
  run.beir.splade-pp-ed.bioasq.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-bioasq.contriever-msmarco \
  --topics beir-v1.0.0-bioasq-test \
  --output run.beir.contriever-msmarco.bioasq.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-bioasq-test \
  run.beir.contriever-msmarco.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-bioasq-test \
  run.beir.contriever-msmarco.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-bioasq-test \
  run.beir.contriever-msmarco.bioasq.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-bioasq.bge-base-en-v1.5 \
  --topics beir-v1.0.0-bioasq-test \
  --output run.beir.bge-base-en-v1.5.bioasq.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-bioasq-test \
  run.beir.bge-base-en-v1.5.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-bioasq-test \
  run.beir.bge-base-en-v1.5.bioasq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-bioasq-test \
  run.beir.bge-base-en-v1.5.bioasq.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-nfcorpus.flat \
  --topics beir-v1.0.0-nfcorpus-test \
  --output run.beir.bm25-flat.nfcorpus.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nfcorpus-test \
  run.beir.bm25-flat.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nfcorpus-test \
  run.beir.bm25-flat.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nfcorpus-test \
  run.beir.bm25-flat.nfcorpus.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-nfcorpus.multifield \
  --topics beir-v1.0.0-nfcorpus-test \
  --output run.beir.bm25-multifield.nfcorpus.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nfcorpus-test \
  run.beir.bm25-multifield.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nfcorpus-test \
  run.beir.bm25-multifield.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nfcorpus-test \
  run.beir.bm25-multifield.nfcorpus.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-nfcorpus.splade-pp-ed \
  --topics beir-v1.0.0-nfcorpus.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.nfcorpus.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nfcorpus-test \
  run.beir.splade-pp-ed.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nfcorpus-test \
  run.beir.splade-pp-ed.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nfcorpus-test \
  run.beir.splade-pp-ed.nfcorpus.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-nfcorpus.contriever-msmarco \
  --topics beir-v1.0.0-nfcorpus-test \
  --output run.beir.contriever-msmarco.nfcorpus.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nfcorpus-test \
  run.beir.contriever-msmarco.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nfcorpus-test \
  run.beir.contriever-msmarco.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nfcorpus-test \
  run.beir.contriever-msmarco.nfcorpus.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-nfcorpus.bge-base-en-v1.5 \
  --topics beir-v1.0.0-nfcorpus-test \
  --output run.beir.bge-base-en-v1.5.nfcorpus.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nfcorpus-test \
  run.beir.bge-base-en-v1.5.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nfcorpus-test \
  run.beir.bge-base-en-v1.5.nfcorpus.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nfcorpus-test \
  run.beir.bge-base-en-v1.5.nfcorpus.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-nq.flat \
  --topics beir-v1.0.0-nq-test \
  --output run.beir.bm25-flat.nq.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nq-test \
  run.beir.bm25-flat.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nq-test \
  run.beir.bm25-flat.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nq-test \
  run.beir.bm25-flat.nq.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-nq.multifield \
  --topics beir-v1.0.0-nq-test \
  --output run.beir.bm25-multifield.nq.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nq-test \
  run.beir.bm25-multifield.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nq-test \
  run.beir.bm25-multifield.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nq-test \
  run.beir.bm25-multifield.nq.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-nq.splade-pp-ed \
  --topics beir-v1.0.0-nq.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.nq.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nq-test \
  run.beir.splade-pp-ed.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nq-test \
  run.beir.splade-pp-ed.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nq-test \
  run.beir.splade-pp-ed.nq.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-nq.contriever-msmarco \
  --topics beir-v1.0.0-nq-test \
  --output run.beir.contriever-msmarco.nq.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nq-test \
  run.beir.contriever-msmarco.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nq-test \
  run.beir.contriever-msmarco.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nq-test \
  run.beir.contriever-msmarco.nq.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-nq.bge-base-en-v1.5 \
  --topics beir-v1.0.0-nq-test \
  --output run.beir.bge-base-en-v1.5.nq.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-nq-test \
  run.beir.bge-base-en-v1.5.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-nq-test \
  run.beir.bge-base-en-v1.5.nq.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-nq-test \
  run.beir.bge-base-en-v1.5.nq.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-hotpotqa.flat \
  --topics beir-v1.0.0-hotpotqa-test \
  --output run.beir.bm25-flat.hotpotqa.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-hotpotqa-test \
  run.beir.bm25-flat.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-hotpotqa-test \
  run.beir.bm25-flat.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-hotpotqa-test \
  run.beir.bm25-flat.hotpotqa.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-hotpotqa.multifield \
  --topics beir-v1.0.0-hotpotqa-test \
  --output run.beir.bm25-multifield.hotpotqa.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-hotpotqa-test \
  run.beir.bm25-multifield.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-hotpotqa-test \
  run.beir.bm25-multifield.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-hotpotqa-test \
  run.beir.bm25-multifield.hotpotqa.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-hotpotqa.splade-pp-ed \
  --topics beir-v1.0.0-hotpotqa.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.hotpotqa.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-hotpotqa-test \
  run.beir.splade-pp-ed.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-hotpotqa-test \
  run.beir.splade-pp-ed.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-hotpotqa-test \
  run.beir.splade-pp-ed.hotpotqa.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-hotpotqa.contriever-msmarco \
  --topics beir-v1.0.0-hotpotqa-test \
  --output run.beir.contriever-msmarco.hotpotqa.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-hotpotqa-test \
  run.beir.contriever-msmarco.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-hotpotqa-test \
  run.beir.contriever-msmarco.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-hotpotqa-test \
  run.beir.contriever-msmarco.hotpotqa.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-hotpotqa.bge-base-en-v1.5 \
  --topics beir-v1.0.0-hotpotqa-test \
  --output run.beir.bge-base-en-v1.5.hotpotqa.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-hotpotqa-test \
  run.beir.bge-base-en-v1.5.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-hotpotqa-test \
  run.beir.bge-base-en-v1.5.hotpotqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-hotpotqa-test \
  run.beir.bge-base-en-v1.5.hotpotqa.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-fiqa.flat \
  --topics beir-v1.0.0-fiqa-test \
  --output run.beir.bm25-flat.fiqa.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fiqa-test \
  run.beir.bm25-flat.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fiqa-test \
  run.beir.bm25-flat.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fiqa-test \
  run.beir.bm25-flat.fiqa.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-fiqa.multifield \
  --topics beir-v1.0.0-fiqa-test \
  --output run.beir.bm25-multifield.fiqa.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fiqa-test \
  run.beir.bm25-multifield.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fiqa-test \
  run.beir.bm25-multifield.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fiqa-test \
  run.beir.bm25-multifield.fiqa.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-fiqa.splade-pp-ed \
  --topics beir-v1.0.0-fiqa.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.fiqa.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fiqa-test \
  run.beir.splade-pp-ed.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fiqa-test \
  run.beir.splade-pp-ed.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fiqa-test \
  run.beir.splade-pp-ed.fiqa.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-fiqa.contriever-msmarco \
  --topics beir-v1.0.0-fiqa-test \
  --output run.beir.contriever-msmarco.fiqa.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fiqa-test \
  run.beir.contriever-msmarco.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fiqa-test \
  run.beir.contriever-msmarco.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fiqa-test \
  run.beir.contriever-msmarco.fiqa.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-fiqa.bge-base-en-v1.5 \
  --topics beir-v1.0.0-fiqa-test \
  --output run.beir.bge-base-en-v1.5.fiqa.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fiqa-test \
  run.beir.bge-base-en-v1.5.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fiqa-test \
  run.beir.bge-base-en-v1.5.fiqa.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fiqa-test \
  run.beir.bge-base-en-v1.5.fiqa.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-signal1m.flat \
  --topics beir-v1.0.0-signal1m-test \
  --output run.beir.bm25-flat.signal1m.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-signal1m-test \
  run.beir.bm25-flat.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-signal1m-test \
  run.beir.bm25-flat.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-signal1m-test \
  run.beir.bm25-flat.signal1m.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-signal1m.multifield \
  --topics beir-v1.0.0-signal1m-test \
  --output run.beir.bm25-multifield.signal1m.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-signal1m-test \
  run.beir.bm25-multifield.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-signal1m-test \
  run.beir.bm25-multifield.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-signal1m-test \
  run.beir.bm25-multifield.signal1m.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-signal1m.splade-pp-ed \
  --topics beir-v1.0.0-signal1m.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.signal1m.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-signal1m-test \
  run.beir.splade-pp-ed.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-signal1m-test \
  run.beir.splade-pp-ed.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-signal1m-test \
  run.beir.splade-pp-ed.signal1m.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-signal1m.contriever-msmarco \
  --topics beir-v1.0.0-signal1m-test \
  --output run.beir.contriever-msmarco.signal1m.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-signal1m-test \
  run.beir.contriever-msmarco.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-signal1m-test \
  run.beir.contriever-msmarco.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-signal1m-test \
  run.beir.contriever-msmarco.signal1m.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-signal1m.bge-base-en-v1.5 \
  --topics beir-v1.0.0-signal1m-test \
  --output run.beir.bge-base-en-v1.5.signal1m.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-signal1m-test \
  run.beir.bge-base-en-v1.5.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-signal1m-test \
  run.beir.bge-base-en-v1.5.signal1m.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-signal1m-test \
  run.beir.bge-base-en-v1.5.signal1m.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-trec-news.flat \
  --topics beir-v1.0.0-trec-news-test \
  --output run.beir.bm25-flat.trec-news.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-news-test \
  run.beir.bm25-flat.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-news-test \
  run.beir.bm25-flat.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-news-test \
  run.beir.bm25-flat.trec-news.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-trec-news.multifield \
  --topics beir-v1.0.0-trec-news-test \
  --output run.beir.bm25-multifield.trec-news.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-news-test \
  run.beir.bm25-multifield.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-news-test \
  run.beir.bm25-multifield.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-news-test \
  run.beir.bm25-multifield.trec-news.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-trec-news.splade-pp-ed \
  --topics beir-v1.0.0-trec-news.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.trec-news.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-news-test \
  run.beir.splade-pp-ed.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-news-test \
  run.beir.splade-pp-ed.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-news-test \
  run.beir.splade-pp-ed.trec-news.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-trec-news.contriever-msmarco \
  --topics beir-v1.0.0-trec-news-test \
  --output run.beir.contriever-msmarco.trec-news.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-news-test \
  run.beir.contriever-msmarco.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-news-test \
  run.beir.contriever-msmarco.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-news-test \
  run.beir.contriever-msmarco.trec-news.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-trec-news.bge-base-en-v1.5 \
  --topics beir-v1.0.0-trec-news-test \
  --output run.beir.bge-base-en-v1.5.trec-news.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-trec-news-test \
  run.beir.bge-base-en-v1.5.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-trec-news-test \
  run.beir.bge-base-en-v1.5.trec-news.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-trec-news-test \
  run.beir.bge-base-en-v1.5.trec-news.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-robust04.flat \
  --topics beir-v1.0.0-robust04-test \
  --output run.beir.bm25-flat.robust04.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-robust04-test \
  run.beir.bm25-flat.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-robust04-test \
  run.beir.bm25-flat.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-robust04-test \
  run.beir.bm25-flat.robust04.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-robust04.multifield \
  --topics beir-v1.0.0-robust04-test \
  --output run.beir.bm25-multifield.robust04.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-robust04-test \
  run.beir.bm25-multifield.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-robust04-test \
  run.beir.bm25-multifield.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-robust04-test \
  run.beir.bm25-multifield.robust04.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-robust04.splade-pp-ed \
  --topics beir-v1.0.0-robust04.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.robust04.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-robust04-test \
  run.beir.splade-pp-ed.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-robust04-test \
  run.beir.splade-pp-ed.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-robust04-test \
  run.beir.splade-pp-ed.robust04.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-robust04.contriever-msmarco \
  --topics beir-v1.0.0-robust04-test \
  --output run.beir.contriever-msmarco.robust04.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-robust04-test \
  run.beir.contriever-msmarco.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-robust04-test \
  run.beir.contriever-msmarco.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-robust04-test \
  run.beir.contriever-msmarco.robust04.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-robust04.bge-base-en-v1.5 \
  --topics beir-v1.0.0-robust04-test \
  --output run.beir.bge-base-en-v1.5.robust04.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-robust04-test \
  run.beir.bge-base-en-v1.5.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-robust04-test \
  run.beir.bge-base-en-v1.5.robust04.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-robust04-test \
  run.beir.bge-base-en-v1.5.robust04.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-arguana.flat \
  --topics beir-v1.0.0-arguana-test \
  --output run.beir.bm25-flat.arguana.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-arguana-test \
  run.beir.bm25-flat.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-arguana-test \
  run.beir.bm25-flat.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-arguana-test \
  run.beir.bm25-flat.arguana.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-arguana.multifield \
  --topics beir-v1.0.0-arguana-test \
  --output run.beir.bm25-multifield.arguana.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-arguana-test \
  run.beir.bm25-multifield.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-arguana-test \
  run.beir.bm25-multifield.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-arguana-test \
  run.beir.bm25-multifield.arguana.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-arguana.splade-pp-ed \
  --topics beir-v1.0.0-arguana.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.arguana.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-arguana-test \
  run.beir.splade-pp-ed.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-arguana-test \
  run.beir.splade-pp-ed.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-arguana-test \
  run.beir.splade-pp-ed.arguana.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-arguana.contriever-msmarco \
  --topics beir-v1.0.0-arguana-test \
  --output run.beir.contriever-msmarco.arguana.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-arguana-test \
  run.beir.contriever-msmarco.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-arguana-test \
  run.beir.contriever-msmarco.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-arguana-test \
  run.beir.contriever-msmarco.arguana.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "" \
  --index beir-v1.0.0-arguana.bge-base-en-v1.5 \
  --topics beir-v1.0.0-arguana-test \
  --output run.beir.bge-base-en-v1.5.arguana.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-arguana-test \
  run.beir.bge-base-en-v1.5.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-arguana-test \
  run.beir.bge-base-en-v1.5.arguana.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-arguana-test \
  run.beir.bge-base-en-v1.5.arguana.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-webis-touche2020.flat \
  --topics beir-v1.0.0-webis-touche2020-test \
  --output run.beir.bm25-flat.webis-touche2020.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-webis-touche2020-test \
  run.beir.bm25-flat.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-webis-touche2020-test \
  run.beir.bm25-flat.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-webis-touche2020-test \
  run.beir.bm25-flat.webis-touche2020.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-webis-touche2020.multifield \
  --topics beir-v1.0.0-webis-touche2020-test \
  --output run.beir.bm25-multifield.webis-touche2020.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-webis-touche2020-test \
  run.beir.bm25-multifield.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-webis-touche2020-test \
  run.beir.bm25-multifield.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-webis-touche2020-test \
  run.beir.bm25-multifield.webis-touche2020.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-webis-touche2020.splade-pp-ed \
  --topics beir-v1.0.0-webis-touche2020.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.webis-touche2020.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-webis-touche2020-test \
  run.beir.splade-pp-ed.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-webis-touche2020-test \
  run.beir.splade-pp-ed.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-webis-touche2020-test \
  run.beir.splade-pp-ed.webis-touche2020.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-webis-touche2020.contriever-msmarco \
  --topics beir-v1.0.0-webis-touche2020-test \
  --output run.beir.contriever-msmarco.webis-touche2020.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-webis-touche2020-test \
  run.beir.contriever-msmarco.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-webis-touche2020-test \
  run.beir.contriever-msmarco.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-webis-touche2020-test \
  run.beir.contriever-msmarco.webis-touche2020.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-webis-touche2020.bge-base-en-v1.5 \
  --topics beir-v1.0.0-webis-touche2020-test \
  --output run.beir.bge-base-en-v1.5.webis-touche2020.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-webis-touche2020-test \
  run.beir.bge-base-en-v1.5.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-webis-touche2020-test \
  run.beir.bge-base-en-v1.5.webis-touche2020.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-webis-touche2020-test \
  run.beir.bge-base-en-v1.5.webis-touche2020.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-android.flat \
  --topics beir-v1.0.0-cqadupstack-android-test \
  --output run.beir.bm25-flat.cqadupstack-android.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bm25-flat.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bm25-flat.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bm25-flat.cqadupstack-android.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-android.multifield \
  --topics beir-v1.0.0-cqadupstack-android-test \
  --output run.beir.bm25-multifield.cqadupstack-android.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bm25-multifield.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bm25-multifield.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bm25-multifield.cqadupstack-android.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-android.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-android.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-android.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-android-test \
  run.beir.splade-pp-ed.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-android-test \
  run.beir.splade-pp-ed.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-android-test \
  run.beir.splade-pp-ed.cqadupstack-android.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-android.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-android-test \
  --output run.beir.contriever-msmarco.cqadupstack-android.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-android-test \
  run.beir.contriever-msmarco.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-android-test \
  run.beir.contriever-msmarco.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-android-test \
  run.beir.contriever-msmarco.cqadupstack-android.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-android.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-android-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-android.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bge-base-en-v1.5.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bge-base-en-v1.5.cqadupstack-android.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-android-test \
  run.beir.bge-base-en-v1.5.cqadupstack-android.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-english.flat \
  --topics beir-v1.0.0-cqadupstack-english-test \
  --output run.beir.bm25-flat.cqadupstack-english.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bm25-flat.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bm25-flat.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bm25-flat.cqadupstack-english.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-english.multifield \
  --topics beir-v1.0.0-cqadupstack-english-test \
  --output run.beir.bm25-multifield.cqadupstack-english.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bm25-multifield.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bm25-multifield.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bm25-multifield.cqadupstack-english.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-english.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-english.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-english.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-english-test \
  run.beir.splade-pp-ed.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-english-test \
  run.beir.splade-pp-ed.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-english-test \
  run.beir.splade-pp-ed.cqadupstack-english.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-english.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-english-test \
  --output run.beir.contriever-msmarco.cqadupstack-english.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-english-test \
  run.beir.contriever-msmarco.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-english-test \
  run.beir.contriever-msmarco.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-english-test \
  run.beir.contriever-msmarco.cqadupstack-english.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-english.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-english-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-english.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bge-base-en-v1.5.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bge-base-en-v1.5.cqadupstack-english.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-english-test \
  run.beir.bge-base-en-v1.5.cqadupstack-english.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-gaming.flat \
  --topics beir-v1.0.0-cqadupstack-gaming-test \
  --output run.beir.bm25-flat.cqadupstack-gaming.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bm25-flat.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bm25-flat.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bm25-flat.cqadupstack-gaming.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-gaming.multifield \
  --topics beir-v1.0.0-cqadupstack-gaming-test \
  --output run.beir.bm25-multifield.cqadupstack-gaming.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bm25-multifield.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bm25-multifield.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bm25-multifield.cqadupstack-gaming.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-gaming.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-gaming.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-gaming.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.splade-pp-ed.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.splade-pp-ed.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.splade-pp-ed.cqadupstack-gaming.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-gaming.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-gaming-test \
  --output run.beir.contriever-msmarco.cqadupstack-gaming.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.contriever-msmarco.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.contriever-msmarco.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.contriever-msmarco.cqadupstack-gaming.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-gaming.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-gaming-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-gaming.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bge-base-en-v1.5.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bge-base-en-v1.5.cqadupstack-gaming.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gaming-test \
  run.beir.bge-base-en-v1.5.cqadupstack-gaming.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-gis.flat \
  --topics beir-v1.0.0-cqadupstack-gis-test \
  --output run.beir.bm25-flat.cqadupstack-gis.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bm25-flat.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bm25-flat.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bm25-flat.cqadupstack-gis.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-gis.multifield \
  --topics beir-v1.0.0-cqadupstack-gis-test \
  --output run.beir.bm25-multifield.cqadupstack-gis.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bm25-multifield.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bm25-multifield.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bm25-multifield.cqadupstack-gis.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-gis.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-gis.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-gis.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.splade-pp-ed.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.splade-pp-ed.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.splade-pp-ed.cqadupstack-gis.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-gis.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-gis-test \
  --output run.beir.contriever-msmarco.cqadupstack-gis.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.contriever-msmarco.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.contriever-msmarco.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.contriever-msmarco.cqadupstack-gis.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-gis.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-gis-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-gis.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bge-base-en-v1.5.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bge-base-en-v1.5.cqadupstack-gis.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-gis-test \
  run.beir.bge-base-en-v1.5.cqadupstack-gis.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-mathematica.flat \
  --topics beir-v1.0.0-cqadupstack-mathematica-test \
  --output run.beir.bm25-flat.cqadupstack-mathematica.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bm25-flat.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bm25-flat.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bm25-flat.cqadupstack-mathematica.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-mathematica.multifield \
  --topics beir-v1.0.0-cqadupstack-mathematica-test \
  --output run.beir.bm25-multifield.cqadupstack-mathematica.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bm25-multifield.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bm25-multifield.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bm25-multifield.cqadupstack-mathematica.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-mathematica.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-mathematica.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-mathematica.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.splade-pp-ed.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.splade-pp-ed.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.splade-pp-ed.cqadupstack-mathematica.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-mathematica.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-mathematica-test \
  --output run.beir.contriever-msmarco.cqadupstack-mathematica.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.contriever-msmarco.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.contriever-msmarco.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.contriever-msmarco.cqadupstack-mathematica.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-mathematica.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-mathematica-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-mathematica.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bge-base-en-v1.5.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bge-base-en-v1.5.cqadupstack-mathematica.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-mathematica-test \
  run.beir.bge-base-en-v1.5.cqadupstack-mathematica.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-physics.flat \
  --topics beir-v1.0.0-cqadupstack-physics-test \
  --output run.beir.bm25-flat.cqadupstack-physics.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bm25-flat.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bm25-flat.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bm25-flat.cqadupstack-physics.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-physics.multifield \
  --topics beir-v1.0.0-cqadupstack-physics-test \
  --output run.beir.bm25-multifield.cqadupstack-physics.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bm25-multifield.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bm25-multifield.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bm25-multifield.cqadupstack-physics.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-physics.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-physics.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-physics.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.splade-pp-ed.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.splade-pp-ed.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.splade-pp-ed.cqadupstack-physics.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-physics.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-physics-test \
  --output run.beir.contriever-msmarco.cqadupstack-physics.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.contriever-msmarco.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.contriever-msmarco.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.contriever-msmarco.cqadupstack-physics.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-physics.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-physics-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-physics.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bge-base-en-v1.5.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bge-base-en-v1.5.cqadupstack-physics.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-physics-test \
  run.beir.bge-base-en-v1.5.cqadupstack-physics.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-programmers.flat \
  --topics beir-v1.0.0-cqadupstack-programmers-test \
  --output run.beir.bm25-flat.cqadupstack-programmers.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bm25-flat.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bm25-flat.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bm25-flat.cqadupstack-programmers.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-programmers.multifield \
  --topics beir-v1.0.0-cqadupstack-programmers-test \
  --output run.beir.bm25-multifield.cqadupstack-programmers.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bm25-multifield.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bm25-multifield.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bm25-multifield.cqadupstack-programmers.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-programmers.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-programmers.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-programmers.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.splade-pp-ed.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.splade-pp-ed.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.splade-pp-ed.cqadupstack-programmers.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-programmers.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-programmers-test \
  --output run.beir.contriever-msmarco.cqadupstack-programmers.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.contriever-msmarco.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.contriever-msmarco.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.contriever-msmarco.cqadupstack-programmers.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-programmers.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-programmers-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-programmers.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bge-base-en-v1.5.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bge-base-en-v1.5.cqadupstack-programmers.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-programmers-test \
  run.beir.bge-base-en-v1.5.cqadupstack-programmers.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-stats.flat \
  --topics beir-v1.0.0-cqadupstack-stats-test \
  --output run.beir.bm25-flat.cqadupstack-stats.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bm25-flat.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bm25-flat.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bm25-flat.cqadupstack-stats.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-stats.multifield \
  --topics beir-v1.0.0-cqadupstack-stats-test \
  --output run.beir.bm25-multifield.cqadupstack-stats.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bm25-multifield.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bm25-multifield.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bm25-multifield.cqadupstack-stats.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-stats.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-stats.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-stats.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.splade-pp-ed.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.splade-pp-ed.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.splade-pp-ed.cqadupstack-stats.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-stats.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-stats-test \
  --output run.beir.contriever-msmarco.cqadupstack-stats.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.contriever-msmarco.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.contriever-msmarco.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.contriever-msmarco.cqadupstack-stats.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-stats.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-stats-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-stats.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bge-base-en-v1.5.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bge-base-en-v1.5.cqadupstack-stats.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-stats-test \
  run.beir.bge-base-en-v1.5.cqadupstack-stats.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-tex.flat \
  --topics beir-v1.0.0-cqadupstack-tex-test \
  --output run.beir.bm25-flat.cqadupstack-tex.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bm25-flat.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bm25-flat.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bm25-flat.cqadupstack-tex.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-tex.multifield \
  --topics beir-v1.0.0-cqadupstack-tex-test \
  --output run.beir.bm25-multifield.cqadupstack-tex.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bm25-multifield.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bm25-multifield.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bm25-multifield.cqadupstack-tex.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-tex.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-tex.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-tex.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.splade-pp-ed.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.splade-pp-ed.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.splade-pp-ed.cqadupstack-tex.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-tex.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-tex-test \
  --output run.beir.contriever-msmarco.cqadupstack-tex.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.contriever-msmarco.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.contriever-msmarco.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.contriever-msmarco.cqadupstack-tex.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-tex.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-tex-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-tex.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bge-base-en-v1.5.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bge-base-en-v1.5.cqadupstack-tex.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-tex-test \
  run.beir.bge-base-en-v1.5.cqadupstack-tex.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-unix.flat \
  --topics beir-v1.0.0-cqadupstack-unix-test \
  --output run.beir.bm25-flat.cqadupstack-unix.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bm25-flat.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bm25-flat.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bm25-flat.cqadupstack-unix.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-unix.multifield \
  --topics beir-v1.0.0-cqadupstack-unix-test \
  --output run.beir.bm25-multifield.cqadupstack-unix.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bm25-multifield.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bm25-multifield.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bm25-multifield.cqadupstack-unix.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-unix.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-unix.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-unix.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.splade-pp-ed.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.splade-pp-ed.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.splade-pp-ed.cqadupstack-unix.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-unix.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-unix-test \
  --output run.beir.contriever-msmarco.cqadupstack-unix.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.contriever-msmarco.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.contriever-msmarco.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.contriever-msmarco.cqadupstack-unix.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-unix.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-unix-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-unix.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bge-base-en-v1.5.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bge-base-en-v1.5.cqadupstack-unix.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-unix-test \
  run.beir.bge-base-en-v1.5.cqadupstack-unix.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-webmasters.flat \
  --topics beir-v1.0.0-cqadupstack-webmasters-test \
  --output run.beir.bm25-flat.cqadupstack-webmasters.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bm25-flat.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bm25-flat.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bm25-flat.cqadupstack-webmasters.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-webmasters.multifield \
  --topics beir-v1.0.0-cqadupstack-webmasters-test \
  --output run.beir.bm25-multifield.cqadupstack-webmasters.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bm25-multifield.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bm25-multifield.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bm25-multifield.cqadupstack-webmasters.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-webmasters.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-webmasters.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-webmasters.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.splade-pp-ed.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.splade-pp-ed.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.splade-pp-ed.cqadupstack-webmasters.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-webmasters.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-webmasters-test \
  --output run.beir.contriever-msmarco.cqadupstack-webmasters.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.contriever-msmarco.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.contriever-msmarco.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.contriever-msmarco.cqadupstack-webmasters.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-webmasters.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-webmasters-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-webmasters.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bge-base-en-v1.5.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bge-base-en-v1.5.cqadupstack-webmasters.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-webmasters-test \
  run.beir.bge-base-en-v1.5.cqadupstack-webmasters.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-wordpress.flat \
  --topics beir-v1.0.0-cqadupstack-wordpress-test \
  --output run.beir.bm25-flat.cqadupstack-wordpress.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bm25-flat.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bm25-flat.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bm25-flat.cqadupstack-wordpress.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-wordpress.multifield \
  --topics beir-v1.0.0-cqadupstack-wordpress-test \
  --output run.beir.bm25-multifield.cqadupstack-wordpress.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bm25-multifield.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bm25-multifield.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bm25-multifield.cqadupstack-wordpress.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-cqadupstack-wordpress.splade-pp-ed \
  --topics beir-v1.0.0-cqadupstack-wordpress.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.cqadupstack-wordpress.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.splade-pp-ed.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.splade-pp-ed.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.splade-pp-ed.cqadupstack-wordpress.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-cqadupstack-wordpress.contriever-msmarco \
  --topics beir-v1.0.0-cqadupstack-wordpress-test \
  --output run.beir.contriever-msmarco.cqadupstack-wordpress.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.contriever-msmarco.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.contriever-msmarco.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.contriever-msmarco.cqadupstack-wordpress.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-cqadupstack-wordpress.bge-base-en-v1.5 \
  --topics beir-v1.0.0-cqadupstack-wordpress-test \
  --output run.beir.bge-base-en-v1.5.cqadupstack-wordpress.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bge-base-en-v1.5.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bge-base-en-v1.5.cqadupstack-wordpress.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-cqadupstack-wordpress-test \
  run.beir.bge-base-en-v1.5.cqadupstack-wordpress.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-quora.flat \
  --topics beir-v1.0.0-quora-test \
  --output run.beir.bm25-flat.quora.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-quora-test \
  run.beir.bm25-flat.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-quora-test \
  run.beir.bm25-flat.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-quora-test \
  run.beir.bm25-flat.quora.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-quora.multifield \
  --topics beir-v1.0.0-quora-test \
  --output run.beir.bm25-multifield.quora.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-quora-test \
  run.beir.bm25-multifield.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-quora-test \
  run.beir.bm25-multifield.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-quora-test \
  run.beir.bm25-multifield.quora.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-quora.splade-pp-ed \
  --topics beir-v1.0.0-quora.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.quora.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-quora-test \
  run.beir.splade-pp-ed.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-quora-test \
  run.beir.splade-pp-ed.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-quora-test \
  run.beir.splade-pp-ed.quora.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-quora.contriever-msmarco \
  --topics beir-v1.0.0-quora-test \
  --output run.beir.contriever-msmarco.quora.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-quora-test \
  run.beir.contriever-msmarco.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-quora-test \
  run.beir.contriever-msmarco.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-quora-test \
  run.beir.contriever-msmarco.quora.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "" \
  --index beir-v1.0.0-quora.bge-base-en-v1.5 \
  --topics beir-v1.0.0-quora-test \
  --output run.beir.bge-base-en-v1.5.quora.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-quora-test \
  run.beir.bge-base-en-v1.5.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-quora-test \
  run.beir.bge-base-en-v1.5.quora.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-quora-test \
  run.beir.bge-base-en-v1.5.quora.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-dbpedia-entity.flat \
  --topics beir-v1.0.0-dbpedia-entity-test \
  --output run.beir.bm25-flat.dbpedia-entity.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bm25-flat.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bm25-flat.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bm25-flat.dbpedia-entity.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-dbpedia-entity.multifield \
  --topics beir-v1.0.0-dbpedia-entity-test \
  --output run.beir.bm25-multifield.dbpedia-entity.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bm25-multifield.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bm25-multifield.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bm25-multifield.dbpedia-entity.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-dbpedia-entity.splade-pp-ed \
  --topics beir-v1.0.0-dbpedia-entity.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.dbpedia-entity.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-dbpedia-entity-test \
  run.beir.splade-pp-ed.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-dbpedia-entity-test \
  run.beir.splade-pp-ed.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-dbpedia-entity-test \
  run.beir.splade-pp-ed.dbpedia-entity.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-dbpedia-entity.contriever-msmarco \
  --topics beir-v1.0.0-dbpedia-entity-test \
  --output run.beir.contriever-msmarco.dbpedia-entity.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-dbpedia-entity-test \
  run.beir.contriever-msmarco.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-dbpedia-entity-test \
  run.beir.contriever-msmarco.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-dbpedia-entity-test \
  run.beir.contriever-msmarco.dbpedia-entity.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-dbpedia-entity.bge-base-en-v1.5 \
  --topics beir-v1.0.0-dbpedia-entity-test \
  --output run.beir.bge-base-en-v1.5.dbpedia-entity.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bge-base-en-v1.5.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bge-base-en-v1.5.dbpedia-entity.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-dbpedia-entity-test \
  run.beir.bge-base-en-v1.5.dbpedia-entity.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-scidocs.flat \
  --topics beir-v1.0.0-scidocs-test \
  --output run.beir.bm25-flat.scidocs.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scidocs-test \
  run.beir.bm25-flat.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scidocs-test \
  run.beir.bm25-flat.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scidocs-test \
  run.beir.bm25-flat.scidocs.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-scidocs.multifield \
  --topics beir-v1.0.0-scidocs-test \
  --output run.beir.bm25-multifield.scidocs.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scidocs-test \
  run.beir.bm25-multifield.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scidocs-test \
  run.beir.bm25-multifield.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scidocs-test \
  run.beir.bm25-multifield.scidocs.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-scidocs.splade-pp-ed \
  --topics beir-v1.0.0-scidocs.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.scidocs.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scidocs-test \
  run.beir.splade-pp-ed.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scidocs-test \
  run.beir.splade-pp-ed.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scidocs-test \
  run.beir.splade-pp-ed.scidocs.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-scidocs.contriever-msmarco \
  --topics beir-v1.0.0-scidocs-test \
  --output run.beir.contriever-msmarco.scidocs.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scidocs-test \
  run.beir.contriever-msmarco.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scidocs-test \
  run.beir.contriever-msmarco.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scidocs-test \
  run.beir.contriever-msmarco.scidocs.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-scidocs.bge-base-en-v1.5 \
  --topics beir-v1.0.0-scidocs-test \
  --output run.beir.bge-base-en-v1.5.scidocs.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scidocs-test \
  run.beir.bge-base-en-v1.5.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scidocs-test \
  run.beir.bge-base-en-v1.5.scidocs.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scidocs-test \
  run.beir.bge-base-en-v1.5.scidocs.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-fever.flat \
  --topics beir-v1.0.0-fever-test \
  --output run.beir.bm25-flat.fever.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fever-test \
  run.beir.bm25-flat.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fever-test \
  run.beir.bm25-flat.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fever-test \
  run.beir.bm25-flat.fever.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-fever.multifield \
  --topics beir-v1.0.0-fever-test \
  --output run.beir.bm25-multifield.fever.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fever-test \
  run.beir.bm25-multifield.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fever-test \
  run.beir.bm25-multifield.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fever-test \
  run.beir.bm25-multifield.fever.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-fever.splade-pp-ed \
  --topics beir-v1.0.0-fever.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.fever.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fever-test \
  run.beir.splade-pp-ed.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fever-test \
  run.beir.splade-pp-ed.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fever-test \
  run.beir.splade-pp-ed.fever.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-fever.contriever-msmarco \
  --topics beir-v1.0.0-fever-test \
  --output run.beir.contriever-msmarco.fever.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fever-test \
  run.beir.contriever-msmarco.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fever-test \
  run.beir.contriever-msmarco.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fever-test \
  run.beir.contriever-msmarco.fever.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-fever.bge-base-en-v1.5 \
  --topics beir-v1.0.0-fever-test \
  --output run.beir.bge-base-en-v1.5.fever.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-fever-test \
  run.beir.bge-base-en-v1.5.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-fever-test \
  run.beir.bge-base-en-v1.5.fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-fever-test \
  run.beir.bge-base-en-v1.5.fever.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-climate-fever.flat \
  --topics beir-v1.0.0-climate-fever-test \
  --output run.beir.bm25-flat.climate-fever.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-climate-fever-test \
  run.beir.bm25-flat.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-climate-fever-test \
  run.beir.bm25-flat.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-climate-fever-test \
  run.beir.bm25-flat.climate-fever.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-climate-fever.multifield \
  --topics beir-v1.0.0-climate-fever-test \
  --output run.beir.bm25-multifield.climate-fever.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-climate-fever-test \
  run.beir.bm25-multifield.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-climate-fever-test \
  run.beir.bm25-multifield.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-climate-fever-test \
  run.beir.bm25-multifield.climate-fever.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-climate-fever.splade-pp-ed \
  --topics beir-v1.0.0-climate-fever.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.climate-fever.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-climate-fever-test \
  run.beir.splade-pp-ed.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-climate-fever-test \
  run.beir.splade-pp-ed.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-climate-fever-test \
  run.beir.splade-pp-ed.climate-fever.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-climate-fever.contriever-msmarco \
  --topics beir-v1.0.0-climate-fever-test \
  --output run.beir.contriever-msmarco.climate-fever.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-climate-fever-test \
  run.beir.contriever-msmarco.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-climate-fever-test \
  run.beir.contriever-msmarco.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-climate-fever-test \
  run.beir.contriever-msmarco.climate-fever.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-climate-fever.bge-base-en-v1.5 \
  --topics beir-v1.0.0-climate-fever-test \
  --output run.beir.bge-base-en-v1.5.climate-fever.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-climate-fever-test \
  run.beir.bge-base-en-v1.5.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-climate-fever-test \
  run.beir.bge-base-en-v1.5.climate-fever.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-climate-fever-test \
  run.beir.bge-base-en-v1.5.climate-fever.txt
Command to generate run:
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-scifact.flat \
  --topics beir-v1.0.0-scifact-test \
  --output run.beir.bm25-flat.scifact.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scifact-test \
  run.beir.bm25-flat.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scifact-test \
  run.beir.bm25-flat.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scifact-test \
  run.beir.bm25-flat.scifact.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-scifact.multifield \
  --topics beir-v1.0.0-scifact-test \
  --output run.beir.bm25-multifield.scifact.txt \
  --output-format trec \
  --hits 1000 --bm25 --remove-query --fields contents=1.0 title=1.0
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scifact-test \
  run.beir.bm25-multifield.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scifact-test \
  run.beir.bm25-multifield.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scifact-test \
  run.beir.bm25-multifield.scifact.txt
Command to generate run:
python -m pyserini.search.lucene \
  --threads 16 --batch-size 128 \
  --index beir-v1.0.0-scifact.splade-pp-ed \
  --topics beir-v1.0.0-scifact.test.splade-pp-ed \
  --output run.beir.splade-pp-ed.scifact.txt \
  --output-format trec \
  --hits 1000 --impact --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scifact-test \
  run.beir.splade-pp-ed.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scifact-test \
  run.beir.splade-pp-ed.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scifact-test \
  run.beir.splade-pp-ed.scifact.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class contriever --encoder facebook/contriever-msmarco \
  --index beir-v1.0.0-scifact.contriever-msmarco \
  --topics beir-v1.0.0-scifact-test \
  --output run.beir.contriever-msmarco.scifact.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scifact-test \
  run.beir.contriever-msmarco.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scifact-test \
  run.beir.contriever-msmarco.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scifact-test \
  run.beir.contriever-msmarco.scifact.txt
Command to generate run:
python -m pyserini.search.faiss \
  --threads 16 --batch-size 512 \
  --encoder-class auto --encoder BAAI/bge-base-en-v1.5 --l2-norm \
  --query-prefix "Represent this sentence for searching relevant passages:" \
  --index beir-v1.0.0-scifact.bge-base-en-v1.5 \
  --topics beir-v1.0.0-scifact-test \
  --output run.beir.bge-base-en-v1.5.scifact.txt \
  --hits 1000 --remove-query
Evaluation commands:
python -m pyserini.eval.trec_eval \
  -c -m ndcg_cut.10 beir-v1.0.0-scifact-test \
  run.beir.bge-base-en-v1.5.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.100 beir-v1.0.0-scifact-test \
  run.beir.bge-base-en-v1.5.scifact.txt

python -m pyserini.eval.trec_eval \
  -c -m recall.1000 beir-v1.0.0-scifact-test \
  run.beir.bge-base-en-v1.5.scifact.txt
Command to generate run: