the_cauldron HuggingFaceM4
Dataset Card for The Cauldron Dataset description The Cauldron is part of the Idefics2 release. It is a massive collection of 50 vision-language datasets (training sets only) that were used for the fine-tuning of the vision-language model Idefics2. Load the dataset To load the dataset, install the library datasets with pip install datasets. Then, from datasets import load_dataset ds = load_dataset("HuggingFaceM4/the_cauldron", "ai2d") to download… See the full description on the dataset page: https://huggingface.co/datasets/HuggingFaceM4/the_cauldron.
- 类型
- dataset
- 下载量
- 268,865
- 点赞
- 552
- 访问
- public
- 文件
- 0
标签
- 多模态数据集
- 视觉语言模型
- 指令微调
- 视觉问答
- 图像理解
- 微调
摘要
The Cauldron是HuggingFaceM4发布的超大规模视觉语言数据集合集，聚合了50个公开的视觉语言训练数据集，专用于对Idefics2视觉语言大模型的指令微调。数据集包含图像与用户/助手文本对，覆盖视觉问答、图表理解、文档解析、数学推理、OCR等多个子任务，每个子数据集提供独立的加载配置。适用于训练多模态大语言模型的指令跟随与图像理解能力。
README
--- dataset_info: - config_name: ai2d features: - name: images sequence: image - name: texts list: - name: user dtype: string - name: assistant dtype: string - name: source dtype: string splits: - name: train num_bytes: 435362437.84770346 num_examples: 2434 download_size: 438136609 dataset_size: 43…