ljleb

Results 256 comments of ljleb

The original txt2img route does not call any script, except for the one explicitly specified in the body of requests. I'll try to see whether it's possible to only load...

Have you checked the webui [wiki](https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Dependencies#required-dependencies)? There seems to be a [link](https://github.com/adang1345/PythonWin7/raw/master/3.10.6/python-3.10.6-amd64-full.exe) to a windows 7 supported version of python 3.10.6: ![image](https://user-images.githubusercontent.com/32277961/221373163-b80fb9f3-07ce-4a1b-896d-bac7cc68d7a9.png) You can find more unofficially supported python versions...

This sounds like a web browser extension is messing up with the page code. Could definitely be something else entirely but can you try disabling all your web browser extensions...

Browser issues with the webui have been a thing for a while, and are not limited to this extension. We could add warnings and guidance in this extension, but I...

You're pretty close! The missing piece is that you have to put the text fields you want to update in the `outputs` array, otherwise gradio won't update any of these...

For some reason, adding these lines: ``` input_image_dims_txt = gr.HTML(value='') input_image.change(fn=lambda img: '0x0', inputs=input_image, outputs=input_image_dims_txt) ``` just under this line: https://github.com/Mikubill/sd-webui-controlnet/blob/fe40f206450aa60105bd146f48f0cbe6f3c1f497/scripts/controlnet.py#L218 ends up creating an infinite loop on image upload?...

> ability to ignore the img2img source image (essentially replicating what you would get out of txt2img) For txt2img batch, I believe there is already an option in the settings...

Maybe we could let users chose between different strategies for handling different-sized batches for each layer/unit. Here are other suggestions from [this](https://github.com/Mikubill/sd-webui-controlnet/issues/243#issuecomment-1447326395) comment: > 1. handle similarly to zip python...

Update: I found a way to reuse the txt2img route code completely, while also monkey patching the original txt2img route to support an additional `controlnet_units` param. Using our own `/controlnet/*2img`...