nmayorga7/chartqa-tables

ChartQA Tables数据集包含从ChartQA数据集中提取的预结构化表格和元数据。这个伴随数据集为关于图表的问题回答提供了视觉和逻辑推理的基准。它包括从图表图像中提取的 structured tables（CSV格式）和论文格式中格式化的 tables，供模型输入使用。该数据集使得模型能够在带有表格的情况下进行评估，从而提高准确性。

The ChartQA Tables dataset contains pre-extracted tables and metadata from the ChartQA dataset by Ahmed Masry et al. This companion dataset provides a benchmark for question answering about charts with visual and logical reasoning, including structured tables in CSV format and formatted tables in the papers format for model input. The dataset enables the with tables evaluation mode for models, which improves accuracy.