ControlNet AttributeError: 'KModel' object has no attribute 'noise_augmentor'

[Melyns]

I'm trying to use Revision mode in Forge ControlNet extension, but i keep getting this error AttributeError: 'KModel' object has no attribute 'noise_augmentor'.

The model CLIP-ViT-bigG.safetensors is located at D:\AI\Forge\webui\models\ControlNetPreprocessor folder (in case this is important information for diagnosis). I verified the hash of the model downloaded on my PC & the one stored at HuggingFace & they match. My only guess is an error in the forge code somewhere? I'd highly appreciate any assistance.

version: f2.0.1v1.10.1-previous-409-gf7ab23b1
python: 3.10.6
torch: 2.4.0+cu124
xformers: N/A

Here's my console log:

ControlNet - INFO - ControlNet Method CLIP-G (Revision) patched.
To load target model KModel
Begin to load 1 model
[Unload] Trying to free 7676.88 MB for cuda:0 with 0 models keep loaded ...
[Unload] Current free memory is 5682.18 MB ...
[Unload] Unload model CLIPVisionModelWithProjection
[Unload] Current free memory is 9201.43 MB ...
[Memory Management] Current Free GPU Memory: 9201.43 MB
[Memory Management] Required Model Memory: 4897.05 MB
[Memory Management] Required Inference Memory: 1024.00 MB
[Memory Management] Estimated Remaining GPU Memory: 3280.38 MB
Moving model(s) has taken 2.88 seconds
  0%|                                                                                           | 0/20 [00:00<?, ?it/s]
Traceback (most recent call last):
  File "D:\AI\Forge\webui\modules_forge\main_thread.py", line 30, in work
    self.result = self.func(*self.args, **self.kwargs)
  File "D:\AI\Forge\webui\modules\txt2img.py", line 112, in txt2img_function
    processed = processing.process_images(p)
  File "D:\AI\Forge\webui\modules\processing.py", line 815, in process_images
    res = process_images_inner(p)
  File "D:\AI\Forge\webui\modules\processing.py", line 958, in process_images_inner
    samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts)
  File "D:\AI\Forge\webui\modules\processing.py", line 1329, in sample
    samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x))
  File "D:\AI\Forge\webui\modules\sd_samplers_kdiffusion.py", line 238, in sample
    samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args=self.sampler_extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs))
  File "D:\AI\Forge\webui\modules\sd_samplers_common.py", line 272, in launch_sampling
    return func()
  File "D:\AI\Forge\webui\modules\sd_samplers_kdiffusion.py", line 238, in <lambda>
    samples = self.launch_sampling(steps, lambda: self.func(self.model_wrap_cfg, x, extra_args=self.sampler_extra_args, disable=False, callback=self.callback_state, **extra_params_kwargs))
  File "D:\AI\Forge\system\python\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "D:\AI\Forge\webui\k_diffusion\sampling.py", line 627, in sample_dpmpp_2m_sde
    denoised = model(x, sigmas[i] * s_in, **extra_args)
  File "D:\AI\Forge\system\python\lib\site-packages\torch\nn\modules\module.py", line 1553, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "D:\AI\Forge\system\python\lib\site-packages\torch\nn\modules\module.py", line 1562, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\AI\Forge\webui\modules\sd_samplers_cfg_denoiser.py", line 186, in forward
    denoised, cond_pred, uncond_pred = sampling_function(self, denoiser_params=denoiser_params, cond_scale=cond_scale, cond_composition=cond_composition)
  File "D:\AI\Forge\webui\backend\sampling\sampling_function.py", line 357, in sampling_function
    model, x, timestep, uncond, cond, cond_scale, model_options, seed = modifier(model, x, timestep, uncond, cond, cond_scale, model_options, seed)
  File "D:\AI\Forge\webui\extensions-builtin\forge_preprocessor_revision\scripts\preprocessor_revision.py", line 10, in revision_conditioning_modifier
    noise_augmentor = model.noise_augmentor
  File "D:\AI\Forge\system\python\lib\site-packages\torch\nn\modules\module.py", line 1729, in __getattr__
    raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")
AttributeError: 'KModel' object has no attribute 'noise_augmentor'
'KModel' object has no attribute 'noise_augmentor'

edit: added more info

[LittleNightmare]

i meet this problem too. in fact, the Revision is working at the first time (just finish downloaded model), but not work after i interrupt my first genertaion. i have try reopen forge but not work.

我也遇到了这个问题。事实上，Revision在刚用的时候正常运行了（刚下载模型），但在我打断我第一次生成后就不工作了。我尝试重开forge但没解决这个问题

[Melyns]

i meet this problem too. in fact, the Revision is working at the first time (just finish downloaded model), but not work after i interrupt my first genertaion. i have try reopen forge but not work.

我也遇到了这个问题。事实上，Revision在刚用的时候正常运行了（刚下载模型），但在我打断我第一次生成后就不工作了。我尝试重开forge但没解决这个问题

Thats strange... so it only works once then you have to restart forge for it to work again? for me the only method that works is if I download Forge zip from release page and not update it. I believe the issue is due to Torch/dependencies breaking forge but i could be wrong