Models: various bag-of-words approaches on segmented documents
This page describes regression experiments for document ranking on the segmented version of 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 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.
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-segmented
Typical indexing command:
target/appassembler/bin/IndexCollection \
-collection MsMarcoV2DocCollection \
-input /path/to/msmarco-v2-doc-segmented \
-generator DefaultLuceneDocumentGenerator \
-index indexes/lucene-index.msmarco-v2-doc-segmented/ \
-threads 24 -storeRaw \
>& logs/log.msmarco-v2-doc-segmented &
The directory /path/to/msmarco-v2-doc-segmented/ 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.
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:
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-v2-doc-segmented/ \
-topics tools/topics-and-qrels/topics.msmarco-v2-doc.dev.txt \
-topicReader TsvInt \
-output runs/run.msmarco-v2-doc-segmented.bm25-default.topics.msmarco-v2-doc.dev.txt \
-bm25 -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 &
target/appassembler/bin/SearchCollection \
-index indexes/lucene-index.msmarco-v2-doc-segmented/ \
-topics tools/topics-and-qrels/topics.msmarco-v2-doc.dev2.txt \
-topicReader TsvInt \
-output runs/run.msmarco-v2-doc-segmented.bm25-default.topics.msmarco-v2-doc.dev2.txt \
-bm25 -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 &
Evaluation can be performed using trec_eval:
target/appassembler/bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev.txt runs/run.msmarco-v2-doc-segmented.bm25-default.topics.msmarco-v2-doc.dev.txt
target/appassembler/bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev.txt runs/run.msmarco-v2-doc-segmented.bm25-default.topics.msmarco-v2-doc.dev.txt
target/appassembler/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-segmented.bm25-default.topics.msmarco-v2-doc.dev.txt
target/appassembler/bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev2.txt runs/run.msmarco-v2-doc-segmented.bm25-default.topics.msmarco-v2-doc.dev2.txt
target/appassembler/bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.msmarco-v2-doc.dev2.txt runs/run.msmarco-v2-doc-segmented.bm25-default.topics.msmarco-v2-doc.dev2.txt
target/appassembler/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-segmented.bm25-default.topics.msmarco-v2-doc.dev2.txt
With the above commands, you should be able to reproduce the following results:
| MAP@100 | BM25 (default) |
|---|---|
| MS MARCO V2 Doc: Dev | 0.1875 |
| MS MARCO V2 Doc: Dev2 | 0.1903 |
| MRR@100 | BM25 (default) |
| MS MARCO V2 Doc: Dev | 0.1896 |
| MS MARCO V2 Doc: Dev2 | 0.1930 |
| R@100 | BM25 (default) |
| MS MARCO V2 Doc: Dev | 0.6555 |
| MS MARCO V2 Doc: Dev2 | 0.6629 |
| R@1000 | BM25 (default) |
| MS MARCO V2 Doc: Dev | 0.8542 |
| MS MARCO V2 Doc: Dev2 | 0.8549 |