lecslab/glosslm-corpus-split-unimorph

--- dataset_info: features: - name: transcription dtype: string - name: glosses dtype: string - name: translation dtype: string - name: glottocode dtype: string - name: id dtype: string - name: source dtype: string - name: metalang_glottocode dtype: string - name: is_segmented dtype: string - name: language dtype: string - name: metalang dtype: string splits: - name: train num_bytes: 93970123 num_examples: 418718 - name: train_ID num_bytes: 25162511 num_examples: 104928 - name: eval_ID num_bytes: 2738251 num_examples: 11138 - name: test_ID num_bytes: 2879162 num_examples: 11940 - name: train_OOD num_bytes: 1824820 num_examples: 7356 - name: eval_OOD num_bytes: 250664 num_examples: 984 - name: test_OOD num_bytes: 241494 num_examples: 972 download_size: 38044153 dataset_size: 127067025 configs: - config_name: default data_files: - split: train path: data/train-* - split: train_ID path: data/train_ID-* - split: eval_ID path: data/eval_ID-* - split: test_ID path: data/test_ID-* - split: train_OOD path: data/train_OOD-* - split: eval_OOD path: data/eval_OOD-* - split: test_OOD path: data/test_OOD-* ---

原始信息汇总

数据集概述

数据集特征

• transcription: 类型为字符串
• glosses: 类型为字符串
• translation: 类型为字符串
• glottocode: 类型为字符串
• id: 类型为字符串
• source: 类型为字符串
• metalang_glottocode: 类型为字符串
• is_segmented: 类型为字符串
• language: 类型为字符串
• metalang: 类型为字符串

数据集划分

• train:
  • 字节数: 93970123
  • 样本数: 418718
• train_ID:
  • 字节数: 25162511
  • 样本数: 104928
• eval_ID:
  • 字节数: 2738251
  • 样本数: 11138
• test_ID:
  • 字节数: 2879162
  • 样本数: 11940
• train_OOD:
  • 字节数: 1824820
  • 样本数: 7356
• eval_OOD:
  • 字节数: 250664
  • 样本数: 984
• test_OOD:
  • 字节数: 241494
  • 样本数: 972

数据集大小

• 下载大小: 38044153 字节
• 数据集总大小: 127067025 字节

配置

• config_name: default
  • 数据文件:
    • train: data/train-*
    • train_ID: data/train_ID-*
    • eval_ID: data/eval_ID-*
    • test_ID: data/test_ID-*
    • train_OOD: data/train_OOD-*
    • eval_OOD: data/eval_OOD-*
    • test_OOD: data/test_OOD-*