PictoViLT/merged_CG_L3_T

该数据集包含文本和图像相关的特征，适用于文本和图像处理任务。数据集分为训练集和测试集，提供了input_ids, attention_mask, token_type_ids, labels等文本处理相关的特征，以及pixel_values和masked_indices等图像处理相关的特征。此外，还包含了一些元数据信息，如chunk_index和logic_name等。

The dataset includes text and image-related features, suitable for text and image processing tasks. It is split into training and test sets, providing features such as input_ids, attention_mask, token_type_ids, labels for text processing, and pixel_values, masked_indices for image processing. Additionally, it contains metadata information like chunk_index and logic_name.