Regressions for the TREC 2019 Deep Learning Track (Passage Ranking Task)

README.md

@@ -74,6 +74,7 @@ Note that these regressions capture the "out of the box" experience, based on [_
 + [Regressions for the MS MARCO Passage Retrieval Task with doc2query expansion](docs/regressions-msmarco-passage-doc2query.md)
 + [Regressions for the MS MARCO Passage Retrieval Task with docTTTTTquery expansion](docs/regressions-msmarco-passage-docTTTTTquery.md)
 + [Regressions for the MS MARCO Document Retrieval](docs/regressions-msmarco-doc.md)
++ [Regressions for the TREC 2019 Deep Learning Track (Passage Ranking Task)](docs/regressions-dl19-passage.md)
 + [Regressions for the TREC 2019 Deep Learning Track (Document Ranking Task)](docs/regressions-dl19-doc.md)
 + [Regressions for NTCIR-8 ACLIA (IR4QA subtask, Monolingual Chinese)](docs/regressions-ntcir8-zh.md)
 + [Regressions for CLEF 2006 Monolingual French](docs/regressions-clef06-fr.md)

docs/regressions-dl19-doc.md

@@ -1,6 +1,6 @@
 # Anserini: Regressions for [DL19 (Document)](https://github.com/microsoft/TREC-2019-Deep-Learning)
 
-This page documents experiments, integrated into Anserini's regression testing framework, for the TREC 2019 Deep Learning Track (Document Ranking Task) on the MS MARCO document collection using relevance judgments from NIST.
+This page describes experiments, integrated into Anserini's regression testing framework, for the TREC 2019 Deep Learning Track (Document Ranking Task) on the MS MARCO document collection using relevance judgments from NIST.
 Note that the NIST relevance judgments provide far more relevant documents 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 document collection, refer to [this page](experiments-msmarco-doc.md).
 
@@ -13,7 +13,7 @@ Typical indexing command:
 
 ```
 nohup sh target/appassembler/bin/IndexCollection -collection TrecCollection -input /path/to/dl19-doc \
- -index lucene-index.msmarco-doc.pos+docvectors+rawdocs -generator LuceneDocumentGenerator -threads 1 \
+ -index lucene-index.msmacro-doc.pos+docvectors+rawdocs -generator LuceneDocumentGenerator -threads 1 \
  -storePositions -storeDocvectors -storeRawDocs >& log.dl19-doc.pos+docvectors+rawdocs &
 ```
 
@@ -24,7 +24,7 @@ For additional details, see explanation of [common indexing options](common-inde
 ## Retrieval
 
 Topics and qrels are stored in [`src/main/resources/topics-and-qrels/`](../src/main/resources/topics-and-qrels/).
-The regression experiments here evaluate on the 43 topics for which NIST has provided judgmen as part of the TREC 2019 Deep Learning Track.
+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](https://trec.nist.gov/data/deep2019.html).
 
 After indexing has completed, you should be able to perform retrieval as follows:

docs/regressions-dl19-passage.md

@@ -0,0 +1,116 @@
+# Anserini: Regressions for [DL19 (Passage)](https://github.com/microsoft/TREC-2019-Deep-Learning)
+
+This page describes experiments, integrated into Anserini's regression testing framework, for the TREC 2019 Deep Learning Track (Passage Ranking Task) on the MS MARCO passage collection using relevance judgments from NIST.
+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](experiments-msmarco-passage.md).
+
+The exact configurations for these regressions are stored in [this YAML file](../src/main/resources/regression/dl19-passage.yaml).
+Note that this page is automatically generated from [this template](../src/main/resources/docgen/templates/dl19-passage.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/dl19-passage \
+ -index lucene-index.msmarco-passage.pos+docvectors+rawdocs -generator LuceneDocumentGenerator -threads 9 \
+ -storePositions -storeDocvectors -storeRawDocs >& log.dl19-passage.pos+docvectors+rawdocs &
+```
+
+The directory `/path/to/msmarco-passage/` should be a directory containing `jsonl` files converted from the official passage collection, which is in `tsv` format.
+[This page](experiments-msmarco-passage.md) explains how to perform this conversion.
+
+For additional details, see explanation of [common indexing options](common-indexing-options.md).
+
+## Retrieval
+
+Topics and qrels are stored in [`src/main/resources/topics-and-qrels/`](../src/main/resources/topics-and-qrels/).
+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](https://trec.nist.gov/data/deep2019.html).
+
+After indexing has completed, you should be able to perform retrieval as follows:
+
+```
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -output run.dl19-passage.bm25-default.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -rm3 -output run.dl19-passage.bm25-default+rm3.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -axiom -axiom.deterministic -rerankCutoff 20 -output run.dl19-passage.bm25-default+ax.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -bm25prf -output run.dl19-passage.bm25-default+prf.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -bm25.k1 0.82 -bm25.b 0.68 -output run.dl19-passage.bm25-tuned.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -bm25.k1 0.82 -bm25.b 0.68 -rm3 -output run.dl19-passage.bm25-tuned+rm3.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -bm25.k1 0.82 -bm25.b 0.68 -axiom -axiom.deterministic -rerankCutoff 20 -output run.dl19-passage.bm25-tuned+ax.topics.dl19-passage.txt &
+
+nohup target/appassembler/bin/SearchCollection -index lucene-index.dl19-passage.pos+docvectors+rawdocs \
+ -topicreader TsvInt -topics src/main/resources/topics-and-qrels/topics.dl19-passage.txt \
+ -bm25 -bm25.k1 0.82 -bm25.b 0.68 -bm25prf -output run.dl19-passage.bm25-tuned+prf.topics.dl19-passage.txt &
+```
+
+Evaluation can be performed using `trec_eval`:
+
+```
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-default.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-default+rm3.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-default+ax.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-default+prf.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-tuned.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-tuned+rm3.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-tuned+ax.topics.dl19-passage.txt
+
+eval/trec_eval.9.0.4/trec_eval -m map -c -m ndcg_cut.10 -c -m recip_rank -c -m recall.100 -c -m recall.1000 -c src/main/resources/topics-and-qrels/qrels.dl19-passage.txt run.dl19-passage.bm25-tuned+prf.topics.dl19-passage.txt
+```
+
+## Effectiveness
+
+With the above commands, you should be able to replicate the following results:
+
+MAP                                     | BM25 (Default)| +RM3      | +Ax       | +PRF      | BM25 (Tuned)| +RM3      | +Ax       | +PRF      |
+:---------------------------------------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|
+[DL19 (Passage)](https://trec.nist.gov/data/deep2019.html)| 0.3773    | 0.4270    | 0.4651    | 0.4533    | 0.3766    | 0.4249    | 0.4722    | 0.4522    |
+
+
+NDCG@10                                 | BM25 (Default)| +RM3      | +Ax       | +PRF      | BM25 (Tuned)| +RM3      | +Ax       | +PRF      |
+:---------------------------------------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|
+[DL19 (Passage)](https://trec.nist.gov/data/deep2019.html)| 0.5058    | 0.5180    | 0.5511    | 0.5372    | 0.4973    | 0.5231    | 0.5461    | 0.5536    |
+
+
+RR                                      | BM25 (Default)| +RM3      | +Ax       | +PRF      | BM25 (Tuned)| +RM3      | +Ax       | +PRF      |
+:---------------------------------------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|
+[DL19 (Passage)](https://trec.nist.gov/data/deep2019.html)| 0.8245    | 0.8167    | 0.7736    | 0.8170    | 0.8457    | 0.8229    | 0.8218    | 0.8178    |
+
+
+R@100                                   | BM25 (Default)| +RM3      | +Ax       | +PRF      | BM25 (Tuned)| +RM3      | +Ax       | +PRF      |
+:---------------------------------------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|
+[DL19 (Passage)](https://trec.nist.gov/data/deep2019.html)| 0.4531    | 0.4761    | 0.4995    | 0.4974    | 0.4603    | 0.4747    | 0.5065    | 0.4969    |
+
+
+R@1000                                  | BM25 (Default)| +RM3      | +Ax       | +PRF      | BM25 (Tuned)| +RM3      | +Ax       | +PRF      |
+:---------------------------------------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|-----------|
+[DL19 (Passage)](https://trec.nist.gov/data/deep2019.html)| 0.7389    | 0.7882    | 0.8129    | 0.7845    | 0.7384    | 0.7762    | 0.8094    | 0.7894    |
+
+The setting "default" refers the default BM25 settings of `k1=0.9`, `b=0.4`, while "tuned" refers to the tuned setting of `k1=0.82`, `b=0.68` (see [this page](experiments-msmarco-passage.md) for more details about tuning).
+

src/main/resources/docgen/templates/dl19-doc.template

@@ -1,6 +1,6 @@
 # Anserini: Regressions for [DL19 (Document)](https://github.com/microsoft/TREC-2019-Deep-Learning)
 
-This page documents experiments, integrated into Anserini's regression testing framework, for the TREC 2019 Deep Learning Track (Document Ranking Task) on the MS MARCO document collection using relevance judgments from NIST.
+This page describes experiments, integrated into Anserini's regression testing framework, for the TREC 2019 Deep Learning Track (Document Ranking Task) on the MS MARCO document collection using relevance judgments from NIST.
 Note that the NIST relevance judgments provide far more relevant documents 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 document collection, refer to [this page](experiments-msmarco-doc.md).
 
@@ -22,7 +22,7 @@ For additional details, see explanation of [common indexing options](common-inde
 ## Retrieval
 
 Topics and qrels are stored in [`src/main/resources/topics-and-qrels/`](../src/main/resources/topics-and-qrels/).
-The regression experiments here evaluate on the 43 topics for which NIST has provided judgmen as part of the TREC 2019 Deep Learning Track.
+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](https://trec.nist.gov/data/deep2019.html).
 
 After indexing has completed, you should be able to perform retrieval as follows:

src/main/resources/docgen/templates/dl19-passage.template

@@ -0,0 +1,48 @@
+# Anserini: Regressions for [DL19 (Passage)](https://github.com/microsoft/TREC-2019-Deep-Learning)
+
+This page describes experiments, integrated into Anserini's regression testing framework, for the TREC 2019 Deep Learning Track (Passage Ranking Task) on the MS MARCO passage collection using relevance judgments from NIST.
+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](experiments-msmarco-passage.md).
+
+The exact configurations for these regressions are stored in [this YAML file](../src/main/resources/regression/dl19-passage.yaml).
+Note that this page is automatically generated from [this template](../src/main/resources/docgen/templates/dl19-passage.template) as part of Anserini's regression pipeline, so do not modify this page directly; modify the template instead.
+
+## Indexing
+
+Typical indexing command:
+
+```
+${index_cmds}
+```
+
+The directory `/path/to/msmarco-passage/` should be a directory containing `jsonl` files converted from the official passage collection, which is in `tsv` format.
+[This page](experiments-msmarco-passage.md) explains how to perform this conversion.
+
+For additional details, see explanation of [common indexing options](common-indexing-options.md).
+
+## Retrieval
+
+Topics and qrels are stored in [`src/main/resources/topics-and-qrels/`](../src/main/resources/topics-and-qrels/).
+The regression experiments here evaluate on the 43 topics for which NIST has provided judgments as part of the TREC 2019 Deep Learning Track.
+The original data can be found [here](https://trec.nist.gov/data/deep2019.html).
+
+After indexing has completed, you should be able to perform retrieval as follows:
+
+```
+${ranking_cmds}
+```
+
+Evaluation can be performed using `trec_eval`:
+
+```
+${eval_cmds}
+```
+
+## Effectiveness
+
+With the above commands, you should be able to replicate the following results:
+
+${effectiveness}
+
+The setting "default" refers the default BM25 settings of `k1=0.9`, `b=0.4`, while "tuned" refers to the tuned setting of `k1=0.82`, `b=0.68` (see [this page](experiments-msmarco-passage.md) for more details about tuning).
+

src/main/resources/regression/dl19-doc.yaml

@@ -1,5 +1,5 @@
 ---
-name: msmarco-doc
+name: msmacro-doc # Note, this is the name of the collection; this way we can share index with the msmarco regressions
 index_command: target/appassembler/bin/IndexCollection
 index_utils_command: target/appassembler/bin/IndexUtils
 search_command: target/appassembler/bin/SearchCollection
@@ -66,7 +66,7 @@ input_roots:
   - /store/      # on orca
   - /scratch2/   # on damiano
 input: collections/msmarco/doc/
-index_path: indexes/lucene-index.msmarco-doc.pos+docvectors+rawdocs
+index_path: lucene-index.msmarco-doc.pos+docvectors+rawdocs
 index_stats:
   documents: 3213835
   documents (non-empty): 3213835