Compatibility with Vladmandic's Auto1111 fork?

[Eljoseto]

Hi! hako-mikan san, I recently moved from AUTO1111 main repo to this fork, as the main branch has been stale for a long time. Vlad's fork has all the recent pull request and is being actively mantained.

That said, I'm having problems runing the SurperMerger on that fork, maybe becaus many options have been reorganized. In particular I'm getting this error:

Traceback (most recent call last):
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\gradio\routes.py", line 394, in run_predict
output = await app.get_blocks().process_api(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\gradio\blocks.py", line 1075, in process_api
result = await self.call_function(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\gradio\blocks.py", line 884, in call_function
prediction = await anyio.to_thread.run_sync(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\anyio_backends_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\anyio_backends_asyncio.py", line 867, in run
result = context.run(func, *args)
File "D:\MachineLearning\VladMandic\extensions\sd-webui-supermerger\scripts\mergers\mergers.py", line 73, in smergegen
usemodelgen(theta_0,model_a,currentmodel)
File "D:\MachineLearning\VladMandic\extensions\sd-webui-supermerger\scripts\mergers\model_util.py", line 798, in usemodelgen
devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16
AttributeError: 'Namespace' object has no attribute 'upcast_sampling'

The start arguments are: --autolaunch --listen --port 7861 --enable-insecure

System Info:

torch: "2.0.0+cu118 autocast half",
gpu: "NVIDIA GeForce RTX 3060 (12Gb) (compute_37) (8, 6)",
cuda: "11.8",
cudnn: 8700
memory: 31.95
xformers: "0.0.17",
accelerate: "0.18.0",
transformers: "4.26.1"
arch: "AMD64",
cpu: "AMD64 Family 23 Model 113 Stepping 0, AuthenticAMD",
system: "Windows",
python: 3.10.11,
gradio: 3.23.0,

[hako-mikan]

may be fixed

[Eljoseto]

Now it works but only the Weight Sum. The rest of the methodes, and the save option give this error:

Traceback (most recent call last):
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\gradio\routes.py", line 394, in run_predict
output = await app.get_blocks().process_api(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\gradio\blocks.py", line 1075, in process_api
result = await self.call_function(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\gradio\blocks.py", line 884, in call_function
prediction = await anyio.to_thread.run_sync(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\anyio_backends_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "D:\MachineLearning\VladMandic\venv\lib\site-packages\anyio_backends_asyncio.py", line 867, in run
result = context.run(func, *args)
File "D:\MachineLearning\VladMandic\extensions\sd-webui-supermerger\scripts\mergers\mergers.py", line 65, in smergegen
result,currentmodel,modelid,theta_0,metadata = smerge(
File "D:\MachineLearning\VladMandic\extensions\sd-webui-supermerger\scripts\mergers\mergers.py", line 174, in smerge
theta_2 = load_model_weights_m(model_c,False,False,save).copy()
File "D:\MachineLearning\VladMandic\extensions\sd-webui-supermerger\scripts\mergers\mergers.py", line 492, in load_model_weights_m
return sd_models.read_state_dict(checkpoint_info.filename,"cuda")
TypeError: read_state_dict() takes 1 positional argument but 2 were given

[hako-mikan]

I see that someone reported a similar error, and I couldn't figure out the cause of the error, but he was using the forked version.
May be fixed.

[Eljoseto]

Thanks You so much! Your last update is working now