Sirawipa/hosxp_ner_code

该数据集主要用于命名实体识别（NER）任务，包含两个特征：words和ner。words是文本中的单词序列，而ner是对应的命名实体标签序列，包含多种实体类型如B-USAGE、I-USAGE、O等。数据集分为训练集和验证集，训练集有1537个样本，验证集有171个样本。数据集的下载大小为53355字节，总大小为598101字节。

This dataset is primarily used for Named Entity Recognition (NER) tasks and contains two features: words and ner. words is a sequence of words in the text, and ner is the corresponding sequence of named entity labels, including various entity types such as B-USAGE, I-USAGE, O, etc. The dataset is divided into a training set and a validation set, with 1537 samples in the training set and 171 samples in the validation set. The download size of the dataset is 53355 bytes, and the total size is 598101 bytes.

原始信息汇总

数据集概述

特征

• words: 字符串序列
• ner: 命名实体识别序列，包含以下类别标签：
  • 0: B-USAGE
  • 1: I-USAGE
  • 2: O
  • 3: B-DOSE
  • 4: B-FREQ
  • 5: I-FREQ
  • 6: B-TIME
  • 7: I-TIME
  • 8: B-UNIT
  • 9: I-UNIT
  • 10: I-DOSE

数据集划分

• train:
  • 样本数量: 1537
  • 字节数: 538383
• validation:
  • 样本数量: 171
  • 字节数: 59718

数据集大小

• 下载大小: 53355 字节
• 数据集总大小: 598101 字节

配置

• default:
  • 训练数据路径: data/train-*
  • 验证数据路径: data/validation-*