error.. size of tensor a (123) must match .... image2image

[Pelicam95]

Greetings, when I try to use the image2image I get the following error :

Iteration: 1/1
Traceback (most recent call last):
File "/usr/local/lib/python3.8/site-packages/gradio/routes.py", line 247, in run_predict
output = await app.blocks.process_api(
File "/usr/local/lib/python3.8/site-packages/gradio/blocks.py", line 641, in process_api
predictions, duration = await self.call_function(fn_index, processed_input)
File "/usr/local/lib/python3.8/site-packages/gradio/blocks.py", line 556, in call_function
prediction = await anyio.to_thread.run_sync(
File "/usr/local/lib/python3.8/site-packages/anyio/to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "/usr/local/lib/python3.8/site-packages/anyio/_backends/_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "/usr/local/lib/python3.8/site-packages/anyio/_backends/_asyncio.py", line 867, in run
result = context.run(func, *args)
File "/content/stable-diffusion/scripts/webui.py", line 1288, in img2img
output_images, seed, info, stats = process_images(
File "/content/stable-diffusion/scripts/webui.py", line 819, in process_images
samples_ddim = func_sample(init_data=init_data, x=x, conditioning=c, unconditional_conditioning=uc, sampler_name=sampler_name)
File "/content/stable-diffusion/scripts/webui.py", line 1194, in sample
xi = x0 + noise
RuntimeError: The size of tensor a (123) must match the size of tensor b (64) at non-singleton dimension 3
Shutting down...

does anyone know the solution?

[biohack78]

I don't know how to fix it but I think this problem happened after I replaced the 'webui.py' and other files when adding the GFPGAN and RealESRGAN models to SD. I could be completely wrong though.

EDIT: (I don't know how to explain this well) OK I found the issue. For example the image I am using to input into img2img is 640x640. By adjusting the width and height sliders around, you change the value of the second number in the RuntimeError line.

It works no issues when you match the number. --> "The size of tensor a (64) must match the size of tensor b (56) at non-singleton dimension 3"

[Pelicam95]

yes, I had already discovered that, it stopped working overnight so yes, it must have been something they changed, I'm surprised no one mentioned it

ps : my temporary solution is to use the "mask" mode for some reason so if it works.