regression.hc4-neuclir22-zh-en.yaml Maven / Gradle / Ivy

Go to download
---
corpus: neuclir22-zh-en
corpus_path: collections/multilingual/neuclir22-zh-en/

index_path: indexes/lucene-index.neuclir22-zh-en
collection_class: NeuClirCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 8
index_options: -storePositions -storeDocvectors -storeRaw
index_stats:
  documents: 3179206
  documents (non-empty): 3179203

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 (Chinese): test-topic title](https://github.com/hltcoe/HC4)"
    id: test_title
    path: topics.hc4-v1.0-zh.en.test.title.tsv
    qrel: qrels.hc4-neuclir22-zh.test.txt
  - name: "[HC4 (Chinese): test-topic description](https://github.com/hltcoe/HC4)"
    id: test_description
    path: topics.hc4-v1.0-zh.en.test.desc.tsv
    qrel: qrels.hc4-neuclir22-zh.test.txt
  - name: "[HC4 (Chinese): test-topic description+title](https://github.com/hltcoe/HC4)"
    id: test_description_title
    path: topics.hc4-v1.0-zh.en.test.desc.title.tsv
    qrel: qrels.hc4-neuclir22-zh.test.txt

models:
  - name: bm25-default
    display: BM25 (default)
    params: -bm25
    results:
      MAP:
        - 0.0839
        - 0.0773
        - 0.0861
      nDCG@20:
        - 0.1271
        - 0.1217
        - 0.1369
      J@20:
        - 0.0860
        - 0.0800
        - 0.0860
      Recall@1000:
        - 0.5982
        - 0.5573
        - 0.6182
  - name: bm25-default+rm3
    display: +RM3
    params: -bm25 -rm3
    results:
      MAP:
        - 0.0730
        - 0.0845
        - 0.0822
      nDCG@20:
        - 0.1142
        - 0.1250
        - 0.1292
      J@20:
        - 0.0810
        - 0.0770
        - 0.0940
      Recall@1000:
        - 0.6357
        - 0.6077
        - 0.6482
  - name: bm25-default+rocchio
    display: +Rocchio
    params: -bm25 -rocchio
    results:
      MAP:
        - 0.0701
        - 0.0842
        - 0.0865
      nDCG@20:
        - 0.1106
        - 0.1238
        - 0.1327
      J@20:
        - 0.0790
        - 0.0820
        - 0.0870
      Recall@1000:
        - 0.6364
        - 0.6142
        - 0.6516