SwayStar123/imagenet_288_dcae_fp8_captions

这是一个包含图像特征和相应标题的数据集，图像特征通过一个降维模型得到，每个样本都包括一个键、类别、文件名、高度、标签、宽度以及对应的标题。数据集分为训练集，并且提供了默认配置。此外，数据集的标题来源于两个不同的HuggingFace数据集。

This dataset contains image features and corresponding captions, with the image features obtained through a dimensionality reduction model. Each sample includes a key, category, filename, height, label, width, and a matching caption. The dataset is split into a training set and provides a default configuration. Additionally, the captions for the dataset are sourced from two different HuggingFace datasets.