ELEPOT comments

Repositories
Issues
Comments

Results 2 comments of


                                            ELEPOT

Much Worse test result when using gradio_canny2image.py than validation result.

I have the same problem. I trained a controlnet model with my own dataset, and the results are much worse when I load the model in automatic1111's webui than the...

Much Worse test result when using gradio_canny2image.py than validation result.

![image](https://github.com/lllyasviel/ControlNet/assets/109792465/7ce8ab8c-7196-4492-9ef9-1de8af065350) After digging through some code I found that in the evaluation process, the CFG scale is 9, the sampling step is 50, and the sampler seems to be ddim...