lxj616
lxj616
@fiht 谢谢，我的本意是把这些步骤全都自动化起来，因为我曾经打算把gourdscan装进docker容器里，举例来说，我之前打包https://github.com/lxj616/docker-blxr 用户就直接用install.php安装就可以了，但是对于用户而言，要想进到docker容器里修改配置文件就是很费劲的事情了（会是肯定都会，但麻烦的事不一定每人都愿意做），如果能自动化起来，更有利于工具的打包和部署，只是一个对开发者的建议而已，感谢你热心的回复
@hyungkwonko Yes it sure does, I already implemented and tested on my own, need some coding Just pass the x0 and mask into the DDIM sampler, and you can do...
@loboere If you mean prompt using part of real image to create similar image, you can try my PR, which does inpainting such as generate a cat wearing a green...
I am not sure if this could be solved by Text + partial image prompting #34 , but I have implemented #34 and made a PR https://github.com/CompVis/latent-diffusion/pull/57 maybe you could...
@rozentill @Sugarbank I had encountered the same problem in another project (A tensorflow gpu repo), by changing the 'CUDAtensor' to 'FloatTensor' do the trick for me, although the vgg model...
@hxngiee I trained the model using examples/research_projects/dreambooth_inpaint/train_dreambooth_inpaint.py from thedarkzeno or patil-suraj And because we are doing video, load the dataset as (b, c, f, h, w) instead of (b, c,...
@hxngiee It's the text2image model with new temporal layers, the text2image model is the stable diffusion, the new layer needs to be trained similar to dreambooth example, since you ask...
@chavinlo I dropped my messy script at https://gist.github.com/lxj616/5134368f44aca837304530695ee100ea But I bet it would be quicker if you modify the original train_dreambooth.py from diffusers than to debug mine, I barely make...
@chavinlo Thanks for asking but 24GB is enough for testing if I pre-compute the embeddings and save them into webdataset, since I see you got A100 (perhaps 40GB vram), you...
@chavinlo One npz contains all video frames of one single video, the loop is dealing with a batch, and the final example['f8'] is a batch of video frames with shape...