DiffEnc-32-8 trained on ImageNet32

Checkpoints for an image generation model trained on ImageNet32.The model was made in Jax. See the github repository for code to load the checkpoints.The model is a variational diffusion model (VDM, https://arxiv.org/abs/2107.00630) with an added trainable time-dependent encoder trained for the article "DiffEnc: Variational Diffusion with a Learned Encoder" (https://arxiv.org/abs/2310.19789).Model uses v-parametrization for the loss. The diffusion model is of size 32 and the encoder is of size 8. That is, the diffusion model uses 32 "down-blocks" in the U-net. See details in article.Model was trained on ImageNet32 for 1.5 million steps.Random seeds: 1, 2, 13

本数据集包含基于 ImageNet32 训练的图像生成模型的检查点。该模型采用 Jax 编程框架构建。具体加载检查点的代码可查阅 GitHub 仓库。模型为变分扩散模型（VDM，参见 https://arxiv.org/abs/2107.00630），并额外集成了可训练的时间依赖编码器，该编码器专为“DiffEnc：基于学习编码器的变分扩散”一文（参见 https://arxiv.org/abs/2310.19789）而设计。模型采用 v 参数化方法进行损失计算。扩散模型尺寸为 32，编码器尺寸为 8，即扩散模型在 U-net 中使用了 32 个“下采样块”。详细内容请参考相关论文。模型在 ImageNet32 数据集上经过 1.5 百万步的训练。随机种子为 1、2、13。