Skip to content

Latest commit

 

History

History
86 lines (67 loc) · 5.21 KB

regressions-msmarco-v2-doc.md

File metadata and controls

86 lines (67 loc) · 5.21 KB

Anserini Regressions: MS MARCO (V2) Document Ranking

Models: various bag-of-words approaches on complete documents

This page describes regression experiments for document ranking on the MS MARCO (V2) document corpus using the dev queries, which is integrated into Anserini's regression testing framework. Here, we cover bag-of-words baselines. For additional instructions on working with the MS MARCO V2 document corpus, refer to this page.

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 msmarco-v2-doc

Indexing

Typical indexing command:

bin/run.sh io.anserini.index.IndexCollection \
  -threads 24 \
  -collection MsMarcoV2DocCollection \
  -input /path/to/msmarco-v2-doc \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-inverted.msmarco-v2-doc/ \
  -storeRaw \
  >& logs/log.msmarco-v2-doc &

The directory /path/to/msmarco-v2-doc/ should be a directory containing the compressed jsonl files that comprise the corpus. See this page for additional details.

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.msmarco-v2-doc/ \
  -topics tools/topics-and-qrels/topics.msmarco-v2-doc.dev.txt \
  -topicReader TsvInt \
  -output runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev.txt \
  -bm25 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-inverted.msmarco-v2-doc/ \
  -topics tools/topics-and-qrels/topics.msmarco-v2-doc.dev2.txt \
  -topicReader TsvInt \
  -output runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev2.txt \
  -bm25 &

Evaluation can be performed using trec_eval:

bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev.txt runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev.txt
bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev.txt runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev.txt
bin/trec_eval -c -M 100 -m map -c -M 100 -m recip_rank tools/topics-and-qrels/qrels.msmarco-v2-doc.dev.txt runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev2.txt runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev2.txt
bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev2.txt runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev2.txt
bin/trec_eval -c -M 100 -m map -c -M 100 -m recip_rank tools/topics-and-qrels/qrels.msmarco-v2-doc.dev2.txt runs/run.msmarco-v2-doc.bm25-default.topics.msmarco-v2-doc.dev2.txt

Effectiveness

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

MAP@100 BM25 (default)
MS MARCO V2 Doc: Dev 0.1552
MS MARCO V2 Doc: Dev2 0.1639
MRR@100 BM25 (default)
MS MARCO V2 Doc: Dev 0.1572
MS MARCO V2 Doc: Dev2 0.1659
R@100 BM25 (default)
MS MARCO V2 Doc: Dev 0.5956
MS MARCO V2 Doc: Dev2 0.5970
R@1000 BM25 (default)
MS MARCO V2 Doc: Dev 0.8054
MS MARCO V2 Doc: Dev2 0.8029