regression.msmarco-v1-passage.yaml Maven / Gradle / Ivy

Go to download
---
corpus: msmarco-passage
corpus_path: collections/msmarco/passage/

index_path: indexes/lucene-inverted.msmarco-v1-passage/
collection_class: JsonCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 9
index_options: -storePositions -storeDocvectors -storeRaw
index_stats:
  documents: 8841823
  documents (non-empty): 8841823
  total terms: 352316036

metrics:
  - metric: AP@1000
    command: bin/trec_eval
    params: -c -m map
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: RR@10
    command: bin/trec_eval
    params: -c -M 10 -m recip_rank
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - 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: "[MS MARCO Passage: Dev](https://github.com/microsoft/MSMARCO-Passage-Ranking)"
    id: dev
    path: topics.msmarco-passage.dev-subset.txt
    qrel: qrels.msmarco-passage.dev-subset.txt

models:
  - name: bm25-default
    display: BM25 (default)
    params: -bm25
    results:
      AP@1000:
        - 0.1926
      RR@10:
        - 0.1840
      R@100:
        - 0.6578
      R@1000:
        - 0.8526
# PRF regressions are no longer maintained for sparse judgments to reduce running times.
# (commenting out instead of removing; in case these numbers are needed, just uncomment and rerun.)
#
#  - name: bm25-default+rm3
#    display: +RM3
#    params: -bm25 -rm3
#    results:
#      AP@1000:
#        - 0.1663
#      RR@10:
#        - 0.1566
#      R@100:
#        - 0.6538
#      R@1000:
#        - 0.8606
#  - name: bm25-default+rocchio
#    display: +Rocchio
#    params: -bm25 -rocchio
#    results:
#      AP@1000:
#        - 0.1690
#      RR@10:
#        - 0.1595
#      R@100:
#        - 0.6553
#      R@1000:
#        - 0.8620
#  - name: bm25-default+rocchio-neg
#    display: +Rocchio*
#    params: -bm25 -rocchio -rocchio.useNegative -rerankCutoff 1000
#    results:
#      AP@1000:
#        - 0.1676
#      RR@10:
#        - 0.1576
#      R@100:
#        - 0.6559
#      R@1000:
#        - 0.8652
#  - name: bm25-default+ax
#    display: +Ax
#    params: -bm25 -axiom -axiom.deterministic -rerankCutoff 20
#    results:
#      AP@1000:
#        - 0.1625
#      RR@10:
#        - 0.1517
#      R@100:
#        - 0.6556
#      R@1000:
#        - 0.8747
#  - name: bm25-default+prf
#    display: +PRF
#    params: -bm25 -bm25prf
#    results:
#      AP@1000:
#        - 0.1520
#      RR@10:
#        - 0.1421
#      R@100:
#        - 0.6535
#      R@1000:
#        - 0.8537
  - name: bm25-tuned
    display: BM25 (tuned)
    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68
    results:
      AP@1000:
        - 0.1958
      RR@10:
        - 0.1875
      R@100:
        - 0.6701
      R@1000:
        - 0.8573
# PRF regressions are no longer maintained for sparse judgments to reduce running times.
# (commenting out instead of removing; in case these numbers are needed, just uncomment and rerun.)
#
#  - name: bm25-tuned+rm3
#    display: +RM3
#    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68 -rm3
#    results:
#      AP@1000:
#        - 0.1741
#      RR@10:
#        - 0.1646
#      R@100:
#        - 0.6674
#      R@1000:
#        - 0.8704
#  - name: bm25-tuned+rocchio
#    display: +Rocchio
#    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68 -rocchio
#    results:
#      AP@1000:
#        - 0.1777
#      RR@10:
#        - 0.1684
#      R@100:
#        - 0.6706
#      R@1000:
#        - 0.8726
#  - name: bm25-tuned+rocchio-neg
#    display: +Rocchio*
#    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68 -rocchio -rocchio.useNegative -rerankCutoff 1000
#    results:
#      AP@1000:
#        - 0.1762
#      RR@10:
#        - 0.1669
#      R@100:
#        - 0.6748
#      R@1000:
#        - 0.8757
#  - name: bm25-tuned+ax
#    display: +Ax
#    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68 -axiom -axiom.deterministic -rerankCutoff 20
#    results:
#      AP@1000:
#        - 0.1699
#      RR@10:
#        - 0.1594
#      R@100:
#        - 0.6721
#      R@1000:
#        - 0.8809
#  - name: bm25-tuned+prf
#    display: +PRF
#    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68 -bm25prf
#    results:
#      AP@1000:
#        - 0.1582
#      RR@10:
#        - 0.1484
#      R@100:
#        - 0.6589
#      R@1000:
#        - 0.8561