chentasker/NoiseZoo

NoiseZoo是一个针对CelebA数据集的潜在表示和扩散反转噪声的统一集合。它为视觉研究提供了一个“罗塞塔石碑”，将相同的图像集映射到10种不同的最先进的编码器和生成架构中。数据集由10个.npy文件组成，每个文件是一个2D NumPy矩阵，形状为(162770, D)，其中D根据不同的模型而变化。文件存储在data/目录下，所有表示都被展平为每张图像的1D向量。

NoiseZoo is a unified collection of latent representations and diffusion inversion noises for the **CelebA** dataset. It provides a "Rosetta Stone" for vision research, mapping the same image set across 10 different state-of-the-art encoder and generative architectures. The dataset consists of 10 `.npy` files. Each file is a 2D NumPy matrix of shape `(162770, D)`, where each row corresponds to the standard CelebA training image index. All files are located in the `data/` directory. All representations have been **flattened** into 1D vectors per image.