Arabic-Clip/Arabic_3M_5M_ViT-B-16-plus-240
收藏Hugging Face2024-07-18 更新2024-07-22 收录
下载链接:
https://hf-mirror.com/datasets/Arabic-Clip/Arabic_3M_5M_ViT-B-16-plus-240
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含阿拉伯语和英语的文本数据,主要用于训练和验证阿拉伯语-英语的多模态模型。数据集分为训练集和验证集,训练集包含200,000行数据,验证集包含5,000行数据。每行数据包含以下特征:index(索引)、url(来源URL)、en_caption(英语标题)、embeddings_en(英语文本的嵌入表示)、caption_ar(阿拉伯语标题)。
This dataset contains Arabic and English text data, primarily used for training and validating Arabic-English multimodal models. The dataset is divided into a training set and a validation set, with the training set containing 200,000 rows of data and the validation set containing 5,000 rows. Each row of data includes the following features: index (index), url (source URL), en_caption (English caption), embeddings_en (embedding representation of English text), and caption_ar (Arabic caption).
提供机构:
Arabic-Clip
原始信息汇总
数据集概述
数据集名称
- Arabic-Clip/Arabic_3M_5M_ViT-B-16-plus-240
数据集结构
-
训练集
- 特征: [index, url, en_caption, embeddings_en, caption_ar]
- 行数: 200000
-
验证集
- 特征: [index, url, en_caption, embeddings_en, caption_ar]
- 行数: 5000



