Almost-AGI-Diffusion/sdxl

该数据集包含由社区投票为“未解决”的提示生成的图像，意味着模型生成的图像未能充分对应提示。数据集包含多个特征，如提示、类别、挑战、注释、图像、模型名称、种子和点赞数。数据集的分割为train，包含219个样本，总大小为25650684字节。生成图像的脚本使用了DiffusionPipeline和DDIMScheduler，并使用了Stable Diffusion XL模型进行图像生成和优化。

This dataset comprises images generated from prompts voted as "unsolved" by the community, meaning that the model-generated images failed to adequately align with the corresponding prompts. This dataset includes multiple attributes such as prompts, categories, challenges, annotations, images, model names, seeds, and like counts. The dataset is split into the training subset, which contains 219 samples with a total size of 25650684 bytes. The scripts used for image generation adopted DiffusionPipeline and DDIMScheduler, and employed the Stable Diffusion XL model for image generation and optimization.

原始信息汇总

数据集概述

数据集信息

• 特征列表:

  • Prompt: 字符串类型
  • Category: 字符串类型
  • Challenge: 字符串类型
  • Note: 字符串类型
  • images: 图像类型
  • model_name: 字符串类型
  • seed: 整数类型 (int64)
  • upvotes: 整数类型 (int64)

• 数据分割:

  • train:
    • 字节数: 25650684.0
    • 样本数: 219

• 下载大小: 25640015

• 数据集大小: 25650684.0

配置信息

• 配置名称: default
  • 数据文件:
    • 分割: train
    • 路径: data/train-*