regression.dl20-passage.splade-pp-ed.cached.yaml Maven / Gradle / Ivy

Go to download
---
corpus: msmarco-passage-splade-pp-ed
corpus_path: collections/msmarco/msmarco-passage-splade-pp-ed

download_url: https://rgw.cs.uwaterloo.ca/pyserini/data/msmarco-passage-splade-pp-ed.tar
download_checksum: e489133bdc54ee1e7c62a32aa582bc77

index_path: indexes/lucene-inverted.msmarco-v1-passage.splade-pp-ed/
collection_class: JsonVectorCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 16
index_options: -impact -pretokenized -storeDocvectors
index_stats:
  documents: 8841823
  documents (non-empty): 8841823
  total terms: 52376261130

metrics:
  - metric: AP@1000
    command: bin/trec_eval
    params: -m map -c -l 2
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: nDCG@10
    command: bin/trec_eval
    params: -m ndcg_cut.10 -c
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: R@100
    command: bin/trec_eval
    params: -m recall.100 -c -l 2
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: R@1000
    command: bin/trec_eval
    params: -m recall.1000 -c -l 2
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false

topic_reader: TsvInt
topics:
  - name: "[DL20 (Passage)](https://trec.nist.gov/data/deep2020.html)"
    id: dl20
    path: topics.dl20.splade-pp-ed.tsv.gz
    qrel: qrels.dl20-passage.txt

models:
  - name: splade-pp-ed-cached
    display: SPLADE++ CoCondenser-EnsembleDistil
    params: -impact -pretokenized
    results:
      AP@1000:
        - 0.5001
      nDCG@10:
        - 0.7198
      R@100:
        - 0.7653
      R@1000:
        - 0.8995
  - name: splade-pp-ed-cached+rm3
    display: +RM3
    params: -impact -pretokenized -rm3
    results:
      AP@1000:
        - 0.5096
      nDCG@10:
        - 0.7131
      R@100:
        - 0.7553
      R@1000:
        - 0.9046
  - name: splade-pp-ed-cached+rocchio
    display: +Rocchio
    params: -impact -pretokenized -rocchio
    results:
      AP@1000:
        - 0.5084
      nDCG@10:
        - 0.7280
      R@100:
        - 0.7704
      R@1000:
        - 0.9069