DeCoDELab/ZeroToCAD-1M-200k

该数据集是一个包含代码和图像的多模态数据集，主要用于与编程或视觉任务相关的应用。数据集包含194,717个训练示例，每个示例由样本ID（字符串类型）、代码（字符串类型）和图像（图像类型）组成，总大小约为1.5GB。数据以train分割形式组织，下载大小约为1.4GB。具体应用场景（如代码生成、图像标注或跨模态学习）未在README中明确说明，但基于字段结构推断可能涉及代码与图像的关联分析。This dataset is a multimodal dataset containing code and images, primarily designed for applications related to programming or visual tasks. It includes 194,717 training examples, each consisting of a sample ID (string type), code (string type), and an image (image type), with a total size of approximately 1.5GB. The data is organized into a train split, with a download size of about 1.4GB. Specific use cases (such as code generation, image captioning, or cross-modal learning) are not explicitly mentioned in the README, but based on the field structure, it may involve the analysis of associations between code and images.