Xiankang Yu

I think when you use the normal to rescale to [-1, 1], you should restore the result to [0,1] when you use save_image function to save the image from tensor....

here is the modified code of dpm_solver ``` """ copy from https://github.com/LuChengTHU/dpm-solver """ import torch import torch.nn.functional as F import math class NoiseScheduleVP: def __init__( self, schedule='discrete', betas=None, alphas_cumprod=None, continuous_beta_0=0.1,...

fix some bugs, but the out is also not good. 

I said trained on celeba-hq，why you said places2？ I only changed batchsize and dataset，but the result is bad. all patchs are inpainting with the same color.

you may misunderstand my words. I said "train 800000 pics of image" mean use the batch size of 8 run about 100000 iters. I just want to verify the result...