This page documents BM25 regression experiments for MIRACL (v1.0) — Bengali.
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 miracl-v1.0-bn
Typical indexing command:
bin/run.sh io.anserini.index.IndexCollection \
-collection MrTyDiCollection \
-input /path/to/miracl-v1.0-bn \
-generator DefaultLuceneDocumentGenerator \
-index indexes/lucene-index.miracl-v1.0-bn/ \
-threads 1 -storePositions -storeDocvectors -storeRaw -language bn \
>& logs/log.miracl-v1.0-bn &
See this page for more details about the MIRACL corpus. For additional details, see explanation of common indexing options.
After indexing has completed, you should be able to perform retrieval as follows:
bin/run.sh io.anserini.search.SearchCollection \
-index indexes/lucene-index.miracl-v1.0-bn/ \
-topics tools/topics-and-qrels/topics.miracl-v1.0-bn-dev.tsv \
-topicReader TsvInt \
-output runs/run.miracl-v1.0-bn.bm25.topics.miracl-v1.0-bn-dev.txt \
-bm25 -hits 100 -language bn &
Evaluation can be performed using trec_eval:
bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.miracl-v1.0-bn-dev.tsv runs/run.miracl-v1.0-bn.bm25.topics.miracl-v1.0-bn-dev.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.miracl-v1.0-bn-dev.tsv runs/run.miracl-v1.0-bn.bm25.topics.miracl-v1.0-bn-dev.txt
With the above commands, you should be able to reproduce the following results:
| nDCG@10 | BM25 |
|---|---|
| MIRACL (Bengali): dev | 0.5079 |
| R@100 | BM25 |
| MIRACL (Bengali): dev | 0.9088 |