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

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

index_path: indexes/lucene-inverted.msmarco-v1-passage.docTTTTTquery/
collection_class: JsonCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 18
index_options: -storePositions -storeDocvectors -storeRaw
index_stats:
  documents: 8841823
  documents (non-empty): 8841823
  total terms: 1986612263

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:
        - 0.2805
      RR@10:
        - 0.2723
      R@100:
        - 0.8192
      R@1000:
        - 0.9470
# PRF regressions are no longer maintained for sparse judgments to reduce running times.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:
#        - 0.2241
#      RR@10:
#        - 0.2139
#      R@100:
#        - 0.8000
#      R@1000:
#        - 0.9460
#  - name: bm25-default+rocchio
#    display: +Rocchio
#    params: -bm25 -rocchio
#    results:
#      AP@1000:
#        - 0.2260
#      RR@10:
#        - 0.2158
#      R@100:
#        - 0.8019
#      R@1000:
#        - 0.9467
#  - name: bm25-default+rocchio-neg
#    display: +Rocchio*
#    params: -bm25 -rocchio -rocchio.useNegative -rerankCutoff 1000
#    results:
#      AP@1000:
#        - 0.2234
#      RR@10:
#        - 0.2129
#      R@100:
#        - 0.8025
#      R@1000:
#        - 0.9475
  - name: bm25-tuned
    display: BM25 (tuned)
    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68
    results:
      AP@1000:
        - 0.2850
      RR@10:
        - 0.2768
      R@100:
        - 0.8190
      R@1000:
        - 0.9471
# 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:
#        - 0.2260
#      RR@10:
#        - 0.2154
#      R@100:
#        - 0.8030
#      R@1000:
#        - 0.9479
#  - name: bm25-tuned+rocchio
#    display: +Rocchio
#    params: -bm25 -bm25.k1 0.82 -bm25.b 0.68 -rocchio
#    results:
#      AP@1000:
#        - 0.2277
#      RR@10:
#        - 0.2174
#      R@100:
#        - 0.8052
#      R@1000:
#        - 0.9496
#  - 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:
#        - 0.2276
#      RR@10:
#        - 0.2173
#      R@100:
#        - 0.8072
#      R@1000:
#        - 0.9493
  - name: bm25-tuned2
    display: BM25 (tuned2)
    params: -bm25 -bm25.k1 2.18 -bm25.b 0.86
    results:
      AP@1000:
        - 0.2893
      RR@10:
        - 0.2816
      R@100:
        - 0.8277
      R@1000:
        - 0.9506
# 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-tuned2+rm3
#    display: +RM3
#    params: -bm25 -bm25.k1 2.18 -bm25.b 0.86 -rm3
#    results:
#      AP@1000:
#        - 0.2472
#      RR@10:
#        - 0.2382
#      R@100:
#        - 0.8226
#      R@1000:
#        - 0.9528
#  - name: bm25-tuned2+rocchio
#    display: +Rocchio
#    params: -bm25 -bm25.k1 2.18 -bm25.b 0.86 -rocchio
#    results:
#      AP@1000:
#        - 0.2488
#      RR@10:
#        - 0.2395
#      R@100:
#        - 0.8238
#      R@1000:
#        - 0.9535
#  - name: bm25-tuned2+rocchio-neg
#    display: +Rocchio*
#    params: -bm25 -bm25.k1 2.18 -bm25.b 0.86 -rocchio -rocchio.useNegative -rerankCutoff 1000
#    results:
#      AP@1000:
#        - 0.2467
#      RR@10:
#        - 0.2372
#      R@100:
#        - 0.8227
#      R@1000:
#        - 0.9539