IAMJB's picture
IAMJB HF staff
Update README.md
a2622bd verified
|
raw
history blame
2.11 kB
metadata
dataset_info:
  features:
    - name: query_id
      dtype: int64
    - name: pos_id
      dtype: int64
    - name: neg_ids
      sequence: int64
    - name: pos_score
      dtype: float64
    - name: neg_scores
      sequence: float64
  splits:
    - name: M3_full_logits
      num_bytes: 1689600000
      num_examples: 3200000
    - name: M3_full_normalized
      num_bytes: 1689600000
      num_examples: 3200000
    - name: M3_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
    - name: monot5_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
    - name: jpbase_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
    - name: jplarge_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
    - name: jpsmall_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
    - name: colbertv2_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
    - name: M3jp_320k_normalized
      num_bytes: 168960000
      num_examples: 320000
  download_size: 3889910104
  dataset_size: 4561920000
configs:
  - config_name: default
    data_files:
      - split: M3_full_logits
        path: data/M3_full_logits-*
      - split: M3_full_normalized
        path: data/M3_full_normalized-*
      - split: M3_320k_normalized
        path: data/M3_320k_normalized-*
      - split: monot5_320k_normalized
        path: data/monot5_320k_normalized-*
      - split: jpbase_320k_normalized
        path: data/jpbase_320k_normalized-*
      - split: jplarge_320k_normalized
        path: data/jplarge_320k_normalized-*
      - split: jpsmall_320k_normalized
        path: data/jpsmall_320k_normalized-*
      - split: colbertv2_320k_normalized
        path: data/colbertv2_320k_normalized-*
      - split: M3jp_320k_normalized
        path: data/M3jp_320k_normalized-*
@misc{clavié2024jacolbertv25optimisingmultivectorretrievers,
      title={JaColBERTv2.5: Optimising Multi-Vector Retrievers to Create State-of-the-Art Japanese Retrievers with Constrained Resources}, 
      author={Benjamin Clavié},
      year={2024},
      eprint={2407.20750},
      archivePrefix={arXiv},
      primaryClass={cs.IR},
      url={https://arxiv.org/abs/2407.20750}, 
}