jesenzhang

How to generate a batch of size=4 when input 1 image using img2img?

1

I want to generate a batchsize>1 images for one input image with img2img, as the txt2img method, ``` x_t_latent = torch.randn((self.frame_bff_size, 4, self.latent_height, self.latent_width)).to( device=self.device, dtype=self.dtype ) ``` x_t_latent can...

Update PhotoMakerNode.py

1

fix the image output for ComfyUI Node

Why don't they look very similar？

3

Try the demo, use image from github, https://huggingface.co/spaces/TencentARC/PhotoMaker ,but it didn't look very similar. 

lib compile failed

2

ubuntu 20.04 + pytorch2.0.1 installed torch-coding 1.2.1 from pip install when run a script in scripts folder ,raise an error incluing this message "gpu/lib_ssd.cu:22:10: fatal error: THC/THCNumerics.cuh: No such file...

SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generation https://github.com/Xiaojiu-z/SSR_Encoder Like IP-adapter, and more accurate，the code based on diffusers has been released,

Can support in comfy native? https://github.com/Con6924/SPM

Can't get similar image

6

The workflow with the same parameter but different input image , the result for asian is not good。  

Trajectory Consistency Distillation Support

21

Trajectory Consistency Distillation needs a confyui native TCDScheduler https://github.com/jabir-zheng/TCD https://huggingface.co/h1t/TCD-SDXL-LoRA

XY Grid save image grid from folder missing the last image

5

I followed the video https://www.youtube.com/watch?v=81R8mgYhq_k, make the XY Grid as bellow,  but,got this result,missing the last one , did I make any mistake? 

训练开始读图时候就报错，维数不对呢

1