PictoViLT/CG_L1_img_F

这是一个包含文本和图像数据的训练集，其中文本数据由input_ids、attention_mask、token_type_ids和labels等特征表示，图像数据由pixel_values特征表示。数据集还包含了masked_indices和metadata信息，metadata中包含了是否仅包含遮蔽的图像标记、逻辑名称以及是否均匀遮蔽图像的字段。训练集大小为1,095,181,943字节，共有617个示例。

This is a training dataset containing both text and image data, with text data represented by features such as input_ids, attention_mask, token_type_ids, and labels, and image data represented by the pixel_values feature. The dataset also includes masked_indices and metadata information, with metadata containing fields such as whether to include only masked tokens in images, logic name, and whether to mask images equally. The training set is 1,095,181,943 bytes in size and contains 617 examples.