Skip to content

Latest commit

 

History

History
159 lines (116 loc) · 11.3 KB

regressions-dl20-doc-segmented.unicoil-noexp.cached.md

File metadata and controls

159 lines (116 loc) · 11.3 KB

Anserini Regressions: TREC 2020 Deep Learning Track (Document)

Model: uniCOIL (without any expansions) on segmented documents (title/segment encoding)

This page describes regression experiments, integrated into Anserini's regression testing framework, using uniCOIL (without any expansions) on the TREC 2020 Deep Learning Track document ranking task. The uniCOIL model is described in the following paper:

Jimmy Lin and Xueguang Ma. A Few Brief Notes on DeepImpact, COIL, and a Conceptual Framework for Information Retrieval Techniques. arXiv:2106.14807.

The experiments on this page are not actually reported in the paper. However, the model is the same, applied to the MS MARCO segmented document corpus (without any expansions). Retrieval uses MaxP technique, where we select the score of the highest-scoring passage from a document as the score for that document to produce a document ranking.

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 then run bin/build.sh 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 dl20-doc-segmented.unicoil-noexp.cached

We make available a version of the MS MARCO document corpus that has already been processed with uniCOIL, i.e., we have performed model inference on every document and stored the output sparse vectors. Thus, no neural inference is involved.

From any machine, the following command will download the corpus and perform the complete regression, end to end:

python src/main/python/run_regression.py --download --index --verify --search --regression dl20-doc-segmented.unicoil-noexp.cached

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-doc-segmented-unicoil-noexp.tar -P collections/
tar xvf collections/msmarco-doc-segmented-unicoil-noexp.tar -C collections/

To confirm, msmarco-doc-segmented-unicoil-noexp.tar is 11 GB and has MD5 checksum 11b226e1cacd9c8ae0a660fd14cdd710. With the corpus downloaded, the following command will perform the remaining steps below:

python src/main/python/run_regression.py --index --verify --search --regression dl20-doc-segmented.unicoil-noexp.cached \
  --corpus-path collections/msmarco-doc-segmented-unicoil-noexp

Indexing

Sample indexing command:

bin/run.sh io.anserini.index.IndexCollection \
  -threads 16 \
  -collection JsonVectorCollection \
  -input /path/to/msmarco-doc-segmented-unicoil-noexp \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-inverted.msmarco-v1-doc-segmented.unicoil-noexp/ \
  -impact -pretokenized -storeDocvectors \
  >& logs/log.msmarco-doc-segmented-unicoil-noexp &

The directory /path/to/msmarco-doc-segmented-unicoil-noexp/ should point to the corpus downloaded above.

The important indexing options to note here are -impact -pretokenized: the first tells Anserini not to encode BM25 doclengths into Lucene's norms (which is the default) and the second option says not to apply any additional tokenization on the uniCOIL tokens. Upon completion, we should have an index with 20,545,677 documents.

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 45 topics for which NIST has provided judgments as part of the TREC 2020 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-doc-segmented.unicoil-noexp/ \
  -topics tools/topics-and-qrels/topics.dl20.unicoil-noexp.0shot.tsv.gz \
  -topicReader TsvInt \
  -output runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached.topics.dl20.unicoil-noexp.0shot.txt \
  -impact -pretokenized -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-inverted.msmarco-v1-doc-segmented.unicoil-noexp/ \
  -topics tools/topics-and-qrels/topics.dl20.unicoil-noexp.0shot.tsv.gz \
  -topicReader TsvInt \
  -output runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rm3.topics.dl20.unicoil-noexp.0shot.txt \
  -impact -pretokenized -rm3 -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-inverted.msmarco-v1-doc-segmented.unicoil-noexp/ \
  -topics tools/topics-and-qrels/topics.dl20.unicoil-noexp.0shot.tsv.gz \
  -topicReader TsvInt \
  -output runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rocchio.topics.dl20.unicoil-noexp.0shot.txt \
  -impact -pretokenized -rocchio -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 &

Evaluation can be performed using trec_eval:

bin/trec_eval -c -M 100 -m map tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached.topics.dl20.unicoil-noexp.0shot.txt

bin/trec_eval -c -M 100 -m map tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rm3.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rm3.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rm3.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rm3.topics.dl20.unicoil-noexp.0shot.txt

bin/trec_eval -c -M 100 -m map tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rocchio.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m ndcg_cut.10 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rocchio.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m recall.100 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rocchio.topics.dl20.unicoil-noexp.0shot.txt
bin/trec_eval -c -m recall.1000 tools/topics-and-qrels/qrels.dl20-doc.txt runs/run.msmarco-doc-segmented-unicoil-noexp.unicoil-noexp-cached+rocchio.topics.dl20.unicoil-noexp.0shot.txt

Effectiveness

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

AP@100 uniCOIL w/ doc2query-T5 expansion +RM3 +Rocchio
DL20 (Doc) 0.3698 0.4140 0.4222
nDCG@10 uniCOIL w/ doc2query-T5 expansion +RM3 +Rocchio
DL20 (Doc) 0.5893 0.6174 0.6181
R@100 uniCOIL w/ doc2query-T5 expansion +RM3 +Rocchio
DL20 (Doc) 0.5872 0.6381 0.6437
R@1000 uniCOIL w/ doc2query-T5 expansion +RM3 +Rocchio
DL20 (Doc) 0.7623 0.8117 0.8151

Note that in the official evaluation for document ranking, all runs were truncated to top-100 hits per query (whereas all top-1000 hits per query were retained for passage ranking). Thus, average precision is computed to depth 100 (i.e., AP@100); nDCG@10 remains unaffected. Remember that we keep qrels of all relevance grades, unlike the case for passage ranking, where relevance grade 1 needs to be discarded when computing certain metrics. Here, we retrieve 1000 hits per query, but measure AP at cutoff 100 (e.g., AP@100). Thus, the experimental results reported here are directly comparable to the results reported in the track overview paper.

Additional Notes

Note that due to MaxP and the need to generate runs to different depths, we can set -hits and -selectMaxPassage.hits differently. The reasonable settings are:

  • -hits 10000 -selectMaxPassage.hits 1000 (as above)
  • -hits 10000 -selectMaxPassage.hits 100
  • -hits 1000 -selectMaxPassage.hits 100

However, for these topics, we get the same effectiveness results; that is, the tie-breaking affects do not manifest in different scores.

Reproduction Log*

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