regression.hc4-v1.0-fa.yaml Maven / Gradle / Ivy

Go to download
---
corpus: hc4-v1.0-fa
corpus_path: collections/multilingual/hc4-v1.0-fa/

index_path: indexes/lucene-index.hc4-v1.0-fa
collection_class: NeuClirCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 8
index_options: -storePositions -storeDocvectors -storeRaw -language fa
index_stats:
  documents: 486486
  documents (non-empty): 486486

metrics:
  - metric: MAP
    command: bin/trec_eval
    params: -c -m map
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: nDCG@20
    command: bin/trec_eval
    params: -c -m ndcg_cut.20
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: J@20
    command: python -m pyserini.eval.trec_eval
    params: -c -m judged.20
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false
  - metric: Recall@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: "[HC4 (Persian): dev-topic title](https://github.com/hltcoe/HC4)"
    id: dev_title
    path: topics.hc4-v1.0-fa.dev.title.tsv
    qrel: qrels.hc4-v1.0-fa.dev.txt
  - name: "[HC4 (Persian): dev-topic description](https://github.com/hltcoe/HC4)"
    id: dev_description
    path: topics.hc4-v1.0-fa.dev.desc.tsv
    qrel: qrels.hc4-v1.0-fa.dev.txt
  - name: "[HC4 (Persian): dev-topic description+title](https://github.com/hltcoe/HC4)"
    id: dev_description_title
    path: topics.hc4-v1.0-fa.dev.desc.title.tsv
    qrel: qrels.hc4-v1.0-fa.dev.txt
  - name: "[HC4 (Persian): test-topic title](https://github.com/hltcoe/HC4)"
    id: test_title
    path: topics.hc4-v1.0-fa.test.title.tsv
    qrel: qrels.hc4-v1.0-fa.test.txt
  - name: "[HC4 (Persian): test-topic description](https://github.com/hltcoe/HC4)"
    id: test_description
    path: topics.hc4-v1.0-fa.test.desc.tsv
    qrel: qrels.hc4-v1.0-fa.test.txt
  - name: "[HC4 (Persian): test-topic description+title](https://github.com/hltcoe/HC4)"
    id: test_description_title
    path: topics.hc4-v1.0-fa.test.desc.title.tsv
    qrel: qrels.hc4-v1.0-fa.test.txt


models:
  - name: bm25-default
    display: BM25 (default)
    params: -bm25 -language fa
    results:
      MAP:
        - 0.2971
        - 0.3243
        - 0.3366
        - 0.2877
        - 0.2928
        - 0.3285
      nDCG@20:
        - 0.3445
        - 0.3475
        - 0.3665
        - 0.3846
        - 0.4039
        - 0.4489
      J@20:
        - 0.4100
        - 0.4750
        - 0.4700
        - 0.4010
        - 0.3890
        - 0.4420
      Recall@1000:
        - 0.7794
        - 0.8491
        - 0.9129
        - 0.8223
        - 0.8402
        - 0.8651
  - name: bm25-default+rm3
    display: +RM3
    params: -bm25 -rm3 -language fa
    results:
      MAP:
        - 0.2865
        - 0.3402
        - 0.2919
        - 0.2954
        - 0.2807
        - 0.3042
      nDCG@20:
        - 0.3446
        - 0.3862
        - 0.3454
        - 0.3812
        - 0.3755
        - 0.4049
      J@20:
        - 0.3300
        - 0.3500
        - 0.3450
        - 0.3800
        - 0.3590
        - 0.3800
      Recall@1000:
        - 0.7683
        - 0.7737
        - 0.8102
        - 0.7755
        - 0.7487
        - 0.8055
  - name: bm25-default+rocchio
    display: +Rocchio
    params: -bm25 -rocchio -language fa
    results:
      MAP:
        - 0.3030
        - 0.3721
        - 0.3881
        - 0.2954
        - 0.2928
        - 0.3274
      nDCG@20:
        - 0.3161
        - 0.3895
        - 0.3962
        - 0.3861
        - 0.3811
        - 0.4256
      J@20:
        - 0.3950
        - 0.5100
        - 0.5250
        - 0.4350
        - 0.4300
        - 0.4580
      Recall@1000:
        - 0.8039
        - 0.8838
        - 0.9154
        - 0.8560
        - 0.8738
        - 0.8883