DylanJHJ/crux-researchy-kd

--- dataset_info: features: - name: query_id dtype: string - name: query_text dtype: string - name: positive_document_ids list: string - name: negative_document_ids list: string - name: positive_document_scores list: float64 - name: negative_document_scores list: float64 - name: answer dtype: 'null' - name: source dtype: string splits: - name: pos_20.neg_51 num_bytes: 285515650 num_examples: 90000 - name: pos_20.neg_51.filtered num_bytes: 272594194 num_examples: 89666 - name: pos_high.neg_zero num_bytes: 50583513 num_examples: 51330 - name: pos_high.neg_low num_bytes: 53039197 num_examples: 51330 - name: pos_high.neg_quarter num_bytes: 66386993 num_examples: 51330 - name: pos_high.neg_half num_bytes: 127087017 num_examples: 51330 - name: pos_half.neg_zero num_bytes: 86257982 num_examples: 81531 - name: pos_low.neg_zero num_bytes: 105015069 num_examples: 84849 - name: pos_zero.neg_high num_bytes: 66273148 num_examples: 63798 - name: train num_bytes: 285515650 num_examples: 90000 download_size: 1050944546 dataset_size: 1398268413 configs: - config_name: default data_files: - split: pos_20.neg_51 path: data/pos_20.neg_51-* - split: pos_20.neg_51.filtered path: data/pos_20.neg_51.filtered-* - split: pos_high.neg_zero path: data/pos_high.neg_zero-* - split: pos_high.neg_low path: data/pos_high.neg_low-* - split: pos_high.neg_quarter path: data/pos_high.neg_quarter-* - split: pos_high.neg_half path: data/pos_high.neg_half-* - split: pos_half.neg_zero path: data/pos_half.neg_zero-* - split: pos_low.neg_zero path: data/pos_low.neg_zero-* - split: pos_zero.neg_high path: data/pos_zero.neg_high-* - split: train path: data/train-* ---