regression.dl23-doc-segmented-msmarco-v2.1.yaml Maven / Gradle / Ivy

Go to download
---
corpus: msmarco-v2.1-doc-segmented
corpus_path: collections/msmarco/msmarco_v2.1_doc_segmented/

index_path: indexes/lucene-inverted.msmarco-v2.1-doc-segmented/
collection_class: MsMarcoV2DocCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 24
index_options: -storeRaw
index_stats:
  documents: 113520750
  documents (non-empty): 113520750
  total terms: 22707699649

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.txt
    qrel: qrels.dl23-doc-msmarco-v2.1.txt

models:
  - name: bm25-default
    display: BM25 (default)
    params: -bm25 -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000
    results:
      MAP@100:
        - 0.1391
      MRR@100:
        - 0.6519
      nDCG@10:
        - 0.3356
      R@100:
        - 0.3049
      R@1000:
        - 0.5852
  - name: bm25-default+rm3
    display: +RM3
    params: -bm25 -rm3 -collection MsMarcoV2DocCollection -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000
    results:
      MAP@100:
        - 0.1716
      MRR@100:
        - 0.6277
      nDCG@10:
        - 0.3435
      R@100:
        - 0.3255
      R@1000:
        - 0.5895
  - name: bm25-default+rocchio
    display: +Rocchio
    params: -bm25 -rocchio -collection MsMarcoV2DocCollection -hits 10000 -selectMaxPassage -selectMaxPassage.delimiter "#" -selectMaxPassage.hits 1000
    results:
      MAP@100:
        - 0.1722
      MRR@100:
        - 0.6202
      nDCG@10:
        - 0.3464
      R@100:
        - 0.3340
      R@1000:
        - 0.6014