Skip to content

Latest commit

 

History

History
60 lines (41 loc) · 3.27 KB

regressions-msmarco-doc-docTTTTTquery-per-doc.md

File metadata and controls

60 lines (41 loc) · 3.27 KB

Anserini: Regressions for MS MARCO Document Ranking

This page documents regression experiments for the MS MARCO document ranking task with per-document docTTTTTquery document expansion, which is integrated into Anserini's regression testing framework. For more complete instructions on how to run end-to-end experiments, 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.

Indexing

Typical indexing command:

nohup sh target/appassembler/bin/IndexCollection -collection JsonCollection \
 -input /path/to/msmarco-doc-docTTTTTquery-per-doc \
 -index indexes/lucene-index.msmarco-doc-docTTTTTquery-per-doc.pos+docvectors+raw \
 -generator DefaultLuceneDocumentGenerator \
 -threads 1 -storePositions -storeDocvectors -storeRaw \
  >& logs/log.msmarco-doc-docTTTTTquery-per-doc &

The directory /path/to/msmarco-doc-docTTTTTquery-per-doc/ should be a directory containing the official document collection (a single file), in TREC format.

For additional details, see explanation of common indexing options.

Retrieval

Topics and qrels are stored in src/main/resources/topics-and-qrels/. The regression experiments here evaluate on the 5193 dev set questions.

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

nohup target/appassembler/bin/SearchCollection -index indexes/lucene-index.msmarco-doc-docTTTTTquery-per-doc.pos+docvectors+raw \
 -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.msmarco-doc.dev.txt \
 -output runs/run.msmarco-doc-docTTTTTquery-per-doc.bm25-default.topics.msmarco-doc.dev.txt \
 -bm25 &

Evaluation can be performed using trec_eval:

tools/eval/trec_eval.9.0.4/trec_eval -m map -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.msmarco-doc.dev.txt runs/run.msmarco-doc-docTTTTTquery-per-doc.bm25-default.topics.msmarco-doc.dev.txt

Effectiveness

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

MAP BM25 (Default)
MS MARCO Document Ranking: Dev Queries 0.2886
R@1000 BM25 (Default)
MS MARCO Document Ranking: Dev Queries 0.9259

See this page for more details. Note that here we are using trec_eval to evaluate the top 1000 hits for each query; beware, the runs provided by MS MARCO organizers for reranking have only 100 hits per query.