fixie-ai/HeySQuAD_human
收藏资源简介:
--- dataset_info: features: - name: audio dtype: audio - name: transcription dtype: string - name: question dtype: string - name: context dtype: string - name: answers list: - name: answer_start dtype: int64 - name: text dtype: string - name: is_impossible dtype: bool - name: id dtype: string - name: plausible_answers list: - name: answer_start dtype: int64 - name: text dtype: string splits: - name: train num_bytes: 17916946579.1 num_examples: 71990 - name: validation num_bytes: 839070526.292 num_examples: 4158 download_size: 14617561895 dataset_size: 18756017105.392 configs: - config_name: default data_files: - split: train path: data/train-* - split: validation path: data/validation-* --- This is a copy of [yijingwu/HeySQuAD_human](https://huggingface.co/datasets/yijingwu/HeySQuAD_human). The only difference is the number of shards used for storing/retrieving the data.
数据集信息: 特征字段: - 音频(audio):数据类型为音频 - 转录文本(transcription):数据类型为字符串 - 问题(question):数据类型为字符串 - 上下文(context):数据类型为字符串 - 答案(answers):列表类型,包含子字段: - 答案起始位置(answer_start):数据类型为int64 - 答案文本(text):数据类型为字符串 - 是否无解(is_impossible):数据类型为布尔值 - 样本ID(id):数据类型为字符串 - 似真答案(plausible_answers):列表类型,包含子字段: - 答案起始位置(answer_start):数据类型为int64 - 答案文本(text):数据类型为字符串 数据集拆分: - 训练集(train):字节大小17916946579.1,样本数71990 - 验证集(validation):字节大小839070526.292,样本数4158 下载总大小:14617561895字节 数据集总存储大小:18756017105.392字节 配置项: - 配置名称(config_name):默认(default) 数据文件路径: - 拆分集:训练集,路径为data/train-* - 拆分集:验证集,路径为data/validation-* 本数据集为[yijingwu/HeySQuAD_human](https://huggingface.co/datasets/yijingwu/HeySQuAD_human)的副本,二者唯一区别在于存储与检索数据时所使用的分片数量。



