wmw

In your reference readme file [https://github.com/hpcaitech/ColossalAI/tree/d4fb7bfda7a2da5480e1187e8d3e40884b42ba11/examples/images/diffusion#:~:text=Teyvat/train_colossalai_teyvat.yaml-,Inference,logdir/checkpoints/last.ckpt%20%5C%0A%20%20%20%20%2D%2Dconfig%20/path/to/logdir/configs/project.yaml%20%20%5C,-usage%3A%20txt2img.py](url) The output project file does not have any **"target"** 2023-06-15T16-43-07-project.yaml which is requires by txt2img and error occures: ``` raise KeyError("Expected key `target` to...

Another question, do you try to finetune stable diffusion V1.5? In this finetune code, I only see you use sd v2.0.ckpt. If I use v1.5-pruned.ckpt, then the mismatch error also...

In fact, this repo has lots of conflicts and errors. I hope your group carefully checks the whole part.

@zhangvia 我现在在看mosaiai https://github.com/mosaicml/diffusion/blob/main/diffusion/train.py 至少能用！

@zhangvia 老哥，3090上可以跑的。。。单机多卡跑过

@zhangvia 老哥试了吗？感觉如何？

@tiandiao123 我试了这个新分支（I HAVE TRIED THIS NEW BRANCH），但是你们直接把ema功能给取消了。是遇到bug还没有解决吗？

@zhangvia 确实，就是个半成品，moving average功能直接腰斩了。

@rtqichen Thanks for your work. But The GPU memory is lost when I use import torch.autograd as ag. My code is as follows: ``` for itr in range(1, 5): print('iteration:...

while I also test that, the spherical shape is good. but for more complex geometry shapes..  bad happens:  could you give me some advice for that?