regression.hc4-neuclir22-ru.yaml Maven / Gradle / Ivy

Go to download
---
corpus: neuclir22-ru
corpus_path: collections/multilingual/neuclir22-ru/

index_path: indexes/lucene-index.neuclir22-ru
collection_class: NeuClirCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 8
index_options: -storePositions -storeDocvectors -storeRaw -language ru
index_stats:
  documents: 4627541
  documents (non-empty): 4627532

metrics:
  - metric: MAP
    command: tools/eval/trec_eval.9.0.4/trec_eval
    params: -c -m map
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: nDCG@20
    command: tools/eval/trec_eval.9.0.4/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: tools/eval/trec_eval.9.0.4/trec_eval
    params: -c -m recall.1000
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: false

topic_reader: TsvInt
topic_root: src/main/resources/topics-and-qrels/
qrels_root: src/main/resources/topics-and-qrels/
topics:
  - name: "[HC4 (Russian): test-topic title](https://github.com/hltcoe/HC4)"
    id: test_title
    path: topics.hc4-v1.0-ru.test.title.tsv
    qrel: qrels.hc4-neuclir22-ru.test.txt
  - name: "[HC4 (Russian): test-topic description](https://github.com/hltcoe/HC4)"
    id: test_description
    path: topics.hc4-v1.0-ru.test.desc.tsv
    qrel: qrels.hc4-neuclir22-ru.test.txt
  - name: "[HC4 (Russian): test-topic description+title](https://github.com/hltcoe/HC4)"
    id: test_description_title
    path: topics.hc4-v1.0-ru.test.desc.title.tsv
    qrel: qrels.hc4-neuclir22-ru.test.txt

models:
  - name: bm25-default
    display: BM25 (default)
    params: -bm25 -language ru
    results:
      MAP:
        - 0.1040
        - 0.0963
        - 0.1264
      nDCG@20:
        - 0.1445
        - 0.1495
        - 0.1762
      J@20:
        - 0.0860
        - 0.0790
        - 0.0900
      Recall@1000:
        - 0.6319
        - 0.6640
        - 0.6667
  - name: bm25-default+rm3
    display: +RM3
    params: -bm25 -rm3 -language ru
    results:
      MAP:
        - 0.0854
        - 0.0654
        - 0.0868
      nDCG@20:
        - 0.1335
        - 0.1061
        - 0.1425
      J@20:
        - 0.0760
        - 0.0620
        - 0.0750
      Recall@1000:
        - 0.6125
        - 0.5408
        - 0.6254
  - name: bm25-default+rocchio
    display: +Rocchio
    params: -bm25 -rocchio -language ru
    results:
      MAP:
        - 0.1231
        - 0.0964
        - 0.1314
      nDCG@20:
        - 0.1655
        - 0.1569
        - 0.1875
      J@20:
        - 0.0930
        - 0.0890
        - 0.0980
      Recall@1000:
        - 0.6982
        - 0.6407
        - 0.6810