Skip to content

Latest commit

 

History

History
60 lines (44 loc) · 2.76 KB

regressions-miracl-v1.0-ru.md

File metadata and controls

60 lines (44 loc) · 2.76 KB

Anserini Regressions: MIRACL (v1.0) — Russian

This page documents BM25 regression experiments for MIRACL (v1.0) — Russian.

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-ru

Indexing

Typical indexing command:

bin/run.sh io.anserini.index.IndexCollection \
  -threads 1 \
  -collection MrTyDiCollection \
  -input /path/to/miracl-v1.0-ru \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-index.miracl-v1.0-ru/ \
  -storePositions -storeDocvectors -storeRaw -language ru \
  >& logs/log.miracl-v1.0-ru &

See this page for more details about the MIRACL corpus. For additional details, see explanation of common indexing options.

Retrieval

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-ru/ \
  -topics tools/topics-and-qrels/topics.miracl-v1.0-ru-dev.tsv \
  -topicReader TsvInt \
  -output runs/run.miracl-v1.0-ru.bm25.topics.miracl-v1.0-ru-dev.txt \
  -bm25 -hits 100 -language ru &

Evaluation can be performed using trec_eval:

bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.miracl-v1.0-ru-dev.tsv runs/run.miracl-v1.0-ru.bm25.topics.miracl-v1.0-ru-dev.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.miracl-v1.0-ru-dev.tsv runs/run.miracl-v1.0-ru.bm25.topics.miracl-v1.0-ru-dev.txt

Effectiveness

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

nDCG@10 BM25
MIRACL (Russian): dev 0.3342
R@100 BM25
MIRACL (Russian): dev 0.6614