search

zoubohao / DenoisingDiffusionProbabilityModel-ddpm-

This may be the simplest implement of DDPM. You can directly run Main.py to train the UNet on CIFAR-10 dataset and see the amazing process of denoising.
MIT License
1.42k stars 156 forks source link

Transfer learning #16

Open ONobody opened 1 year ago

ONobody commented 1 year ago

Hello, how can I use the weight you gave me to fine-tune my training data set? Thank you. @zoubohao

xuzekai1997 commented 1 year ago

First ,change the config to load your dataset .And as long as you load your own dataset，you just need to train model.