Skip to content

bright-leetcode.md

Latest commit

 

History

History
75 lines (55 loc) · 3.78 KB

bright-leetcode.md

File metadata and controls

75 lines (55 loc) · 3.78 KB

Anserini Regressions: BRIGHT — LeetCode

This page documents BM25 regression experiments for BRIGHT — LeetCode.

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 and build Anserini to rebuild the documentation.

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 bright-leetcode

All the BRIGHT corpora are available for download:

wget https://huggingface.co/datasets/castorini/collections-bright/resolve/main/bright-corpus.tar -P collections/
tar xvf collections/bright-corpus.tar -C collections/

The tarball is 284 MB and has MD5 checksum 568b594709a9977369033117bfb6889c. After download and unpacking the corpora, the run_regression.py command above should work without any issue.

Indexing

Typical indexing command:

bin/run.sh io.anserini.index.IndexCollection \
  -threads 1 \
  -collection JsonCollection \
  -input /path/to/bright-leetcode \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-inverted.bright-leetcode/ \
  -storePositions -storeDocvectors -storeRaw \
  >& logs/log.bright-leetcode &

The path /path/to/bright-leetcode/ should point to the corpus downloaded above. 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:

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-inverted.bright-leetcode/ \
  -topics tools/topics-and-qrels/topics.bright-leetcode.tsv.gz \
  -topicReader TsvString \
  -output runs/run.bright-leetcode.bm25.topics.bright-leetcode.txt \
  -bm25 -removeQuery -hits 1000 &

Evaluation can be performed using trec_eval:

bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.bright-leetcode.txt runs/run.bright-leetcode.bm25.topics.bright-leetcode.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.bright-leetcode.txt runs/run.bright-leetcode.bm25.topics.bright-leetcode.txt
bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.bright-leetcode.txt runs/run.bright-leetcode.bm25.topics.bright-leetcode.txt

Effectiveness

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

nDCG@10 BM25
BRIGHT: LeetCode 0.2471
R@100 BM25
BRIGHT: LeetCode 0.5076
R@1000 BM25
BRIGHT: LeetCode 0.8001