Skip to content

Latest commit

 

History

History
111 lines (79 loc) · 6.56 KB

regressions-dl19-passage.bm25-b8.md

File metadata and controls

111 lines (79 loc) · 6.56 KB

Anserini Regressions: TREC 2019 Deep Learning Track (Passage)

Models: BM25 with quantized weights (8 bits)

This page describes baseline experiments, integrated into Anserini's regression testing framework, on the TREC 2019 Deep Learning Track passage ranking task.

Note that the NIST relevance judgments provide far more relevant passages per topic, unlike the "sparse" judgments provided by Microsoft (these are sometimes called "dense" judgments to emphasize this contrast). For additional instructions on working with MS MARCO passage collection, 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 dl19-passage.bm25-b8

From any machine, the following command will download the corpus (as quantized BM25 weights) and perform the complete regression, end to end:

python src/main/python/run_regression.py --download --index --verify --search --regression dl19-passage.bm25-b8

The run_regression.py script automates the following steps, but if you want to perform each step manually, simply copy/paste from the commands below and you'll obtain the same regression results.

Corpus Download

Download the corpus and unpack into collections/:

wget https://rgw.cs.uwaterloo.ca/JIMMYLIN-bucket0/data/msmarco-passage-bm25-b8.tar -P collections/
tar xvf collections/msmarco-passage-bm25-b8.tar -C collections/

To confirm, msmarco-passage-bm25-b8.tar is 1.2 GB and has MD5 checksum 0a623e2c97ac6b7e814bf1323a97b435. With the corpus downloaded, the following command will perform the remaining steps below:

python src/main/python/run_regression.py --index --verify --search --regression dl19-passage.bm25-b8 \
  --corpus-path collections/msmarco-passage-bm25-b8

Indexing

Typical indexing command:

bin/run.sh io.anserini.index.IndexCollection \
  -threads 9 \
  -collection JsonVectorCollection \
  -input /path/to/msmarco-passage-bm25-b8 \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-inverted.msmarco-v1-passage.bm25-b8/ \
  -impact -pretokenized \
  >& logs/log.msmarco-passage-bm25-b8 &

The directory /path/to/msmarco-passage-bm25-b8/ should be a directory containing jsonl files containing quantized BM25 vectors for every document

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. The regression experiments here evaluate on the 43 topics for which NIST has provided judgments as part of the TREC 2019 Deep Learning Track. The original data can be found here.

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-v1-passage.bm25-b8/ \
  -topics tools/topics-and-qrels/topics.dl19-passage.txt \
  -topicReader TsvInt \
  -output runs/run.msmarco-passage-bm25-b8.bm25-b8.topics.dl19-passage.txt \
  -impact &

Evaluation can be performed using trec_eval:

bin/trec_eval -m map -c -l 2 tools/topics-and-qrels/qrels.dl19-passage.txt runs/run.msmarco-passage-bm25-b8.bm25-b8.topics.dl19-passage.txt
bin/trec_eval -m ndcg_cut.10 -c tools/topics-and-qrels/qrels.dl19-passage.txt runs/run.msmarco-passage-bm25-b8.bm25-b8.topics.dl19-passage.txt
bin/trec_eval -m recall.100 -c -l 2 tools/topics-and-qrels/qrels.dl19-passage.txt runs/run.msmarco-passage-bm25-b8.bm25-b8.topics.dl19-passage.txt
bin/trec_eval -m recall.1000 -c -l 2 tools/topics-and-qrels/qrels.dl19-passage.txt runs/run.msmarco-passage-bm25-b8.bm25-b8.topics.dl19-passage.txt

Effectiveness

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

AP@1000 BM25 (default parameters, quantized 8 bits)
DL19 (Passage) 0.3046
nDCG@10 BM25 (default parameters, quantized 8 bits)
DL19 (Passage) 0.4993
R@100 BM25 (default parameters, quantized 8 bits)
DL19 (Passage) 0.4949
R@1000 BM25 (default parameters, quantized 8 bits)
DL19 (Passage) 0.7639

❗ Retrieval metrics here are computed to depth 1000 hits per query (as opposed to 100 hits per query for document ranking). For computing nDCG, remember that we keep qrels of all relevance grades, whereas for other metrics (e.g., AP), relevance grade 1 is considered not relevant (i.e., use the -l 2 option in trec_eval). The experimental results reported here are directly comparable to the results reported in the track overview paper.

Reproduction Log*

To add to this reproduction log, modify this template and run bin/build.sh to rebuild the documentation.