Skip to content

Latest commit

 

History

History
203 lines (174 loc) · 11.5 KB

regressions-disk12.md

File metadata and controls

203 lines (174 loc) · 11.5 KB

Anserini Regressions: TIPSTER Disks 1 & 2

Models: various bag-of-words approaches

This page describes regressions for ad hoc topics from TREC 1-3, which use TIPSTER Disks 1 & 2. 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 disk12

Indexing

Typical indexing command:

bin/run.sh io.anserini.index.IndexCollection \
  -collection TrecCollection \
  -input /path/to/disk12 \
  -generator DefaultLuceneDocumentGenerator \
  -index indexes/lucene-index.disk12/ \
  -threads 16 -storePositions -storeDocvectors -storeRaw \
  >& logs/log.disk12 &

The directory /path/to/disk12/ should be the root directory of TIPSTER Disks 1 & 2, i.e., ls /path/to/disk12/ should bring up subdirectories like doe, wsj.

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. They are downloaded from NIST:

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

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.51-100.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25.topics.adhoc.51-100.txt \
  -bm25 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.101-150.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25.topics.adhoc.101-150.txt \
  -bm25 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.151-200.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25.topics.adhoc.151-200.txt \
  -bm25 &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.51-100.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25+rm3.topics.adhoc.51-100.txt \
  -bm25 -rm3 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.101-150.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25+rm3.topics.adhoc.101-150.txt \
  -bm25 -rm3 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.151-200.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25+rm3.topics.adhoc.151-200.txt \
  -bm25 -rm3 &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.51-100.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25+ax.topics.adhoc.51-100.txt \
  -bm25 -axiom -axiom.deterministic -rerankCutoff 20 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.101-150.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25+ax.topics.adhoc.101-150.txt \
  -bm25 -axiom -axiom.deterministic -rerankCutoff 20 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.151-200.txt \
  -topicReader Trec \
  -output runs/run.disk12.bm25+ax.topics.adhoc.151-200.txt \
  -bm25 -axiom -axiom.deterministic -rerankCutoff 20 &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.51-100.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql.topics.adhoc.51-100.txt \
  -qld &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.101-150.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql.topics.adhoc.101-150.txt \
  -qld &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.151-200.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql.topics.adhoc.151-200.txt \
  -qld &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.51-100.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql+rm3.topics.adhoc.51-100.txt \
  -qld -rm3 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.101-150.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql+rm3.topics.adhoc.101-150.txt \
  -qld -rm3 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.151-200.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql+rm3.topics.adhoc.151-200.txt \
  -qld -rm3 &

bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.51-100.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql+ax.topics.adhoc.51-100.txt \
  -qld -axiom -axiom.deterministic -rerankCutoff 20 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.101-150.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql+ax.topics.adhoc.101-150.txt \
  -qld -axiom -axiom.deterministic -rerankCutoff 20 &
bin/run.sh io.anserini.search.SearchCollection \
  -index indexes/lucene-index.disk12/ \
  -topics tools/topics-and-qrels/topics.adhoc.151-200.txt \
  -topicReader Trec \
  -output runs/run.disk12.ql+ax.topics.adhoc.151-200.txt \
  -qld -axiom -axiom.deterministic -rerankCutoff 20 &

Evaluation can be performed using trec_eval:

bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.51-100.txt runs/run.disk12.bm25.topics.adhoc.51-100.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.101-150.txt runs/run.disk12.bm25.topics.adhoc.101-150.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.151-200.txt runs/run.disk12.bm25.topics.adhoc.151-200.txt

bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.51-100.txt runs/run.disk12.bm25+rm3.topics.adhoc.51-100.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.101-150.txt runs/run.disk12.bm25+rm3.topics.adhoc.101-150.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.151-200.txt runs/run.disk12.bm25+rm3.topics.adhoc.151-200.txt

bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.51-100.txt runs/run.disk12.bm25+ax.topics.adhoc.51-100.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.101-150.txt runs/run.disk12.bm25+ax.topics.adhoc.101-150.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.151-200.txt runs/run.disk12.bm25+ax.topics.adhoc.151-200.txt

bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.51-100.txt runs/run.disk12.ql.topics.adhoc.51-100.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.101-150.txt runs/run.disk12.ql.topics.adhoc.101-150.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.151-200.txt runs/run.disk12.ql.topics.adhoc.151-200.txt

bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.51-100.txt runs/run.disk12.ql+rm3.topics.adhoc.51-100.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.101-150.txt runs/run.disk12.ql+rm3.topics.adhoc.101-150.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.151-200.txt runs/run.disk12.ql+rm3.topics.adhoc.151-200.txt

bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.51-100.txt runs/run.disk12.ql+ax.topics.adhoc.51-100.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.101-150.txt runs/run.disk12.ql+ax.topics.adhoc.101-150.txt
bin/trec_eval -m map -m P.30 tools/topics-and-qrels/qrels.adhoc.151-200.txt runs/run.disk12.ql+ax.topics.adhoc.151-200.txt

Effectiveness

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

MAP BM25 +RM3 +Ax QL +RM3 +Ax
TREC-1 Ad Hoc Topics 51-100 0.2277 0.2614 0.2648 0.2188 0.2464 0.2502
TREC-2 Ad Hoc Topics 101-150 0.2003 0.2579 0.2698 0.2010 0.2424 0.2596
TREC-3 Ad Hoc Topics 151-200 0.2634 0.3345 0.3407 0.2580 0.3029 0.3129
P30 BM25 +RM3 +Ax QL +RM3 +Ax
TREC-1 Ad Hoc Topics 51-100 0.4540 0.4927 0.5127 0.4553 0.4673 0.4947
TREC-2 Ad Hoc Topics 101-150 0.4253 0.4580 0.4720 0.4193 0.4427 0.4760
TREC-3 Ad Hoc Topics 151-200 0.4860 0.5293 0.5273 0.4753 0.5000 0.5187