Skip to content

Latest commit

 

History

History
78 lines (56 loc) · 4.37 KB

regressions-beir-v1.0.0-fiqa-unicoil-noexp.md

File metadata and controls

78 lines (56 loc) · 4.37 KB

Anserini Regressions: BEIR (v1.0.0) — FiQA-2018

Model: uniCOIL (without any expansions)

This page describes regression experiments, integrated into Anserini's regression testing framework, using uniCOIL (without any expansions) on BEIR (v1.0.0) — FiQA-2018. The uniCOIL model is described in the following paper:

Jimmy Lin and Xueguang Ma. A Few Brief Notes on DeepImpact, COIL, and a Conceptual Framework for Information Retrieval Techniques. arXiv:2106.14807.

The exact configurations for these regressions are stored in this YAML file. Note that this page is automatically generated from this template as part of Anserini's regression pipeline, so do not modify this page directly; modify the template instead.

From one of our Waterloo servers (e.g., orca), the following command will perform the complete regression, end to end:

python src/main/python/run_regression.py --index --verify --search --regression beir-v1.0.0-fiqa-unicoil-noexp

All the BEIR corpora, encoded by the uniCOIL-noexp model, are available for download:

wget https://rgw.cs.uwaterloo.ca/pyserini/data/beir-v1.0.0-unicoil-noexp.tar -P collections/
tar xvf collections/beir-v1.0.0-unicoil-noexp.tar -C collections/

The tarball is 30 GB and has MD5 checksum 4fd04d2af816a6637fc12922cccc8a83. After download and unpacking the corpora, the run_regression.py command above should work without any issue.

Indexing

Typical indexing command:

target/appassembler/bin/IndexCollection \
  -collection JsonVectorCollection \
  -input /path/to/beir-v1.0.0-fiqa-unicoil-noexp \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-index.beir-v1.0.0-fiqa-unicoil-noexp/ \
  -threads 16 -impact -pretokenized \
  >& logs/log.beir-v1.0.0-fiqa-unicoil-noexp &

For additional details, see explanation of common indexing options.

Retrieval

Topics and qrels are stored here, which is linked to the Anserini repo as a submodule.

After indexing has completed, you should be able to perform retrieval as follows:

target/appassembler/bin/SearchCollection \
  -index indexes/lucene-index.beir-v1.0.0-fiqa-unicoil-noexp/ \
  -topics tools/topics-and-qrels/topics.beir-v1.0.0-fiqa.test.unicoil-noexp.tsv.gz \
  -topicReader TsvString \
  -output runs/run.beir-v1.0.0-fiqa-unicoil-noexp.unicoil-noexp.topics.beir-v1.0.0-fiqa.test.unicoil-noexp.txt \
  -impact -pretokenized -removeQuery -hits 1000 &

Evaluation can be performed using trec_eval:

target/appassembler/bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.beir-v1.0.0-fiqa.test.txt runs/run.beir-v1.0.0-fiqa-unicoil-noexp.unicoil-noexp.topics.beir-v1.0.0-fiqa.test.unicoil-noexp.txt
target/appassembler/bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.beir-v1.0.0-fiqa.test.txt runs/run.beir-v1.0.0-fiqa-unicoil-noexp.unicoil-noexp.topics.beir-v1.0.0-fiqa.test.unicoil-noexp.txt
target/appassembler/bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.beir-v1.0.0-fiqa.test.txt runs/run.beir-v1.0.0-fiqa-unicoil-noexp.unicoil-noexp.topics.beir-v1.0.0-fiqa.test.unicoil-noexp.txt

Effectiveness

With the above commands, you should be able to reproduce the following results:

nDCG@10 uniCOIL no expansion
BEIR (v1.0.0): FiQA-2018 0.2888
R@100 uniCOIL no expansion
BEIR (v1.0.0): FiQA-2018 0.5533
R@1000 uniCOIL no expansion
BEIR (v1.0.0): FiQA-2018 0.7465