regression.dl23-doc-segmented.unicoil-0shot-v2.cached.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-inverted.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: bin/trec_eval
    params: -c -M 100 -m map
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: no
  - metric: MRR@100
    command: bin/trec_eval
    params: -c -M 100 -m recip_rank
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: nDCG@10
    command: bin/trec_eval
    params: -c -m ndcg_cut.10
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: R@100
    command: bin/trec_eval
    params: -c -m recall.100
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: R@1000
    command: bin/trec_eval
    params: -c -m recall.1000
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false

topic_reader: TsvInt
topics:
  - name: "[DL23 (Doc)](https://microsoft.github.io/msmarco/TREC-Deep-Learning)"
    id: dl23
    path: topics.dl23.unicoil.0shot.tsv.gz
    qrel: qrels.dl23-doc.txt

models:
  - name: unicoil-0shot-cached
    display: uniCOIL (with doc2query-T5) zero-shot
    params: -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 -impact -pretokenized
    results:
      MAP@100:
        - 0.1554
      MRR@100:
        - 0.7793
      nDCG@10:
        - 0.4149
      R@100:
        - 0.3101
      R@1000:
        - 0.5753
  - name: unicoil-0shot-cached+rm3
    display: +RM3
    params: -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 -impact -pretokenized -rm3 -collection JsonVectorCollection
    results:
      MAP@100:
        - 0.1832
      MRR@100:
        - 0.7931
      nDCG@10:
        - 0.4477
      R@100:
        - 0.3380
      R@1000:
        - 0.6067
  - name: unicoil-0shot-cached+rocchio
    display: +Rocchio
    params: -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000 -impact -pretokenized -rocchio -collection JsonVectorCollection
    results:
      MAP@100:
        - 0.1844
      MRR@100:
        - 0.7943
      nDCG@10:
        - 0.4284
      R@100:
        - 0.3450
      R@1000:
        - 0.6257