Suriring/squad_bigbird

该数据集包含用于自然语言处理任务的输入特征和位置信息。具体特征包括input_ids（int32序列）、attention_mask（int8序列）、start_positions（int64类型）和end_positions（int64类型）。数据集分为训练集和验证集，训练集包含86,821个样本，验证集包含5,928个样本。数据集的下载大小为22,406,426字节，总大小为1,901,725,496字节。

This dataset contains input features and position information for natural language processing tasks. The specific features include input_ids (int32 sequence), attention_mask (int8 sequence), start_positions (int64 type), and end_positions (int64 type). The dataset is divided into a training set and a validation set, with the training set containing 86,821 samples and the validation set containing 5,928 samples. The download size of the dataset is 22,406,426 bytes, and the total size is 1,901,725,496 bytes.

原始信息汇总

数据集概述

数据特征

• input_ids: 序列类型为 int32
• attention_mask: 序列类型为 int8
• start_positions: 数据类型为 int64
• end_positions: 数据类型为 int64

数据分割

• train:
  • 字节数: 1780177784
  • 样本数: 86821
• validation:
  • 字节数: 121547712
  • 样本数: 5928

数据集大小

• 下载大小: 22406426 字节
• 总大小: 1901725496 字节

配置

• config_name: default
  • data_files:
    • train: data/train-*
    • validation: data/validation-*