fire12-hi.yaml

---
corpus: fire12-hi
corpus_path: collections/fire/hindi/hi.docs.2011/

index_path: indexes/lucene-index.fire12-hi/
collection_class: CleanTrecCollection
generator_class: DefaultLuceneDocumentGenerator
index_threads: 16
index_options: -storePositions -storeDocvectors -storeRaw -language hi
index_stats:
  documents: 331599
  documents (non-empty): 331599
  total terms: 57038417

metrics:
  - metric: MAP
    command: target/appassembler/bin/trec_eval
    params: -m map
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: P20
    command: target/appassembler/bin/trec_eval
    params: -m P.20
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true
  - metric: nDCG@20
    command: target/appassembler/bin/trec_eval
    params: -m ndcg_cut.20
    separator: "\t"
    parse_index: 2
    metric_precision: 4
    can_combine: true

topic_reader: Trec
topics:
  - name: "[FIRE 2012 (Monolingual Hindi)](https://github.com/castorini/anserini-tools/tree/master/topics-and-qrels/topics.fire12en.176-225.txt)"
    id: fire12
    path: topics.fire12hi.176-225.txt
    qrel: qrels.fire12hi.176-225.txt

models:
  - name: bm25
    display: BM25
    params: -bm25 -language hi
    results:
      MAP:
        - 0.3867
      P20:
        - 0.4470
      nDCG@20:
        - 0.5310