regression.msmarco-v2-doc-segmented-unicoil-0shot-v2.yaml Maven / Gradle / Ivy

Go to download
---
corpus: msmarco-v2-doc-segmented-unicoil-0shot-v2
corpus_path: collections/msmarco/msmarco_v2_doc_segmented_unicoil_0shot_v2/

download_url: https://rgw.cs.uwaterloo.ca/JIMMYLIN-bucket0/data/msmarco_v2_doc_segmented_unicoil_0shot_v2.tar
download_checksum: c5639748c2cbad0152e10b0ebde3b804
download_corpus: msmarco_v2_doc_segmented_unicoil_0shot_v2

index_path: indexes/lucene-index.msmarco-v2-doc-segmented-unicoil-0shot-v2/
collection_class: JsonVectorCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 24
index_options: -impact -pretokenized -storeRaw
index_stats:
  documents: 124131414
  documents (non-empty): 124131414
  total terms: 1204542769110

metrics:
  - metric: MAP@100
    command: tools/eval/trec_eval.9.0.4/trec_eval
    params: -c -M 100 -m map
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: MRR@100
    command: tools/eval/trec_eval.9.0.4/trec_eval
    params: -c -M 100 -m recip_rank
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: R@100
    command: tools/eval/trec_eval.9.0.4/trec_eval
    params: -c -m recall.100
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: R@1000
    command: tools/eval/trec_eval.9.0.4/trec_eval
    params: -c -m recall.1000
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false

topic_reader: TsvInt
topic_root: src/main/resources/topics-and-qrels/
qrels_root: src/main/resources/topics-and-qrels/
topics:
  - name: "[MS MARCO V2 Doc: Dev](https://microsoft.github.io/msmarco/TREC-Deep-Learning.html)"
    id: dev
    path: topics.msmarco-v2-doc.dev.unicoil.0shot.tsv.gz
    qrel: qrels.msmarco-v2-doc.dev.txt
  - name: "[MS MARCO V2 Doc: Dev2](https://microsoft.github.io/msmarco/TREC-Deep-Learning.html)"
    id: dev2
    path: topics.msmarco-v2-doc.dev2.unicoil.0shot.tsv.gz
    qrel: qrels.msmarco-v2-doc.dev2.txt

models:
  - name: unicoil-0shot
    display: uniCOIL (with doc2query-T5) zero-shot
    params: -parallelism 16 -impact -pretokenized -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000
    results:
      MAP@100:
        - 0.2388
        - 0.2422
      MRR@100:
        - 0.2419
        - 0.2445
      R@100:
        - 0.7791
        - 0.7759
      R@1000:
        - 0.9122
        - 0.9172
  - name: unicoil-0shot+rm3
    display: +RM3
    params: -parallelism 16 -impact -pretokenized -rm3 -collection JsonVectorCollection -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000
    results:
      MAP@100:
        - 0.2174
        - 0.2198
      MRR@100:
        - 0.2197
        - 0.2225
      R@100:
        - 0.7684
        - 0.7640
      R@1000:
        - 0.9175
        - 0.9220
  - name: unicoil-0shot+rocchio
    display: +Rocchio
    params: -parallelism 16 -impact -pretokenized -rocchio -collection JsonVectorCollection -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000
    results:
      MAP@100:
        - 0.2229
        - 0.2200
      MRR@100:
        - 0.2252
        - 0.2225
      R@100:
        - 0.7775
        - 0.7747
      R@1000:
        - 0.9232
        - 0.9253