A1111 Directml AMD GPU - upscale in FSL stopped working after the git pull/update of the extensions #198
Assignees
CalvinBackup (Calvin Backup)
Clear assignees
CalvinBackup (Calvin Backup)
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: CalvinBackup/sd-webui-faceswaplab#198
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Hey,
so yesterday I git pulled the A1111 Directml build together with updating extensions. Since then, FaceSwap Lab face upscale is not working with any of the upscale. I tried to git clone the A1111 repo, build venv from a scratch and get FaceSwap Lab repo as the one and only extension. It didn't help and the faceswap lab generates the following problem when using an upscale:
2024-09-04 21:54:20,554 - FaceSwapLab - INFO - Try to use model : C:\Users\mbord_\Documents\stable-diffusion-webui-directml\models\faceswaplab\inswapper_128.onnx
2024-09-04 21:54:20,872 - FaceSwapLab - INFO - blend all faces together
2024-09-04 21:54:20,872 - FaceSwapLab - INFO - loading face DEFAN_FULL.safetensors
2024-09-04 21:54:20,874 - FaceSwapLab - INFO - Int Gender : 0
2024-09-04 21:54:20,874 - FaceSwapLab - INFO - Process face 0
2024-09-04 21:54:20,884 - FaceSwapLab - INFO - Source Gender 0
2024-09-04 21:54:20,885 - FaceSwapLab - INFO - Target faces count : 1
2024-09-04 21:54:20,885 - FaceSwapLab - INFO - swap face 0
2024-09-04 21:54:21,681 - FaceSwapLab - INFO - ********************************************************************************
2024-09-04 21:54:21,681 - FaceSwapLab - INFO - Inswapper
2024-09-04 21:54:21,683 - FaceSwapLab - INFO - Upscale with ESRGAN_4x scale = 4
tiled upscale: 0%| | 0/1 [00:00<?, ?it/s]
Traceback (most recent call last):
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\upscaled_inswapper.py", line 215, in get
bgr_fake = self.upscale_and_restore(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\upscaled_inswapper.py", line 153, in upscale_and_restore
upscaled = upscaling.upscale_img(pil_img, pp_options)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_postprocessing\upscaling.py", line 19, in upscale_img
result_image = pp_options.upscaler.scaler.upscale(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\upscaler.py", line 68, in upscale
img = self.do_upscale(img, selected_model)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\esrgan_model.py", line 36, in do_upscale
return esrgan_upscale(model, img)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\esrgan_model.py", line 57, in esrgan_upscale
return upscale_with_model(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\upscaler_utils.py", line 74, in upscale_with_model
output = upscale_pil_patch(model, tile)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\upscaler_utils.py", line 48, in upscale_pil_patch
return torch_bgr_to_pil_image(model(tensor))
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\spandrel_helpers\model_descriptor.py", line 472, in call
output = self.call_fn(self.model, image)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\spandrel_helpers\model_descriptor.py", line 439, in
self.call_fn = call_fn or (lambda model, image: model(image))
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1532, in wrapped_call_impl
return self.call_impl(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1541, in call_impl
return forward_call(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\spandrel\architectures\ESRGAN\arch\RRDB.py", line 142, in forward
return self.model(x)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1532, in wrapped_call_impl
return self.call_impl(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1541, in call_impl
return forward_call(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\container.py", line 217, in forward
input = module(input)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1532, in wrapped_call_impl
return self.call_impl(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1541, in call_impl
return forward_call(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\extensions-builtin\Lora\networks.py", line 599, in network_Conv2d_forward
return originals.Conv2d_forward(self, input)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\conv.py", line 460, in forward
return self.conv_forward(input, self.weight, self.bias)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\conv.py", line 456, in conv_forward
return F.conv2d(input, weight, bias, self.stride,
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\modules\dml\amp\autocast_mode.py", line 43, in
setattr(resolved_obj, func_path[-1], lambda *args, **kwargs: forward(op, args, kwargs))
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\modules\dml\amp\autocast_mode.py", line 15, in forward
return op(*args, **kwargs)
RuntimeError: Cannot set version_counter for inference tensor
2024-09-04 21:54:21,948 - FaceSwapLab - ERROR - Conversion failed Cannot set version_counter for inference tensor
2024-09-04 21:54:21,948 - FaceSwapLab - ERROR - Failed to swap face in postprocess method : Cannot set version_counter for inference tensor
Traceback (most recent call last):
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab.py", line 187, in postprocess
swapped_images = swapper.process_images_units(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\swapper.py", line 841, in process_images_units
swapped = process_image_unit(model, units[0], image, info, force_blend)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\swapper.py", line 752, in process_image_unit
result: ImageResult = swap_face(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\swapper.py", line 658, in swap_face
raise e
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\swapper.py", line 646, in swap_face
result = face_swapper.get(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\upscaled_inswapper.py", line 328, in get
raise e
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\upscaled_inswapper.py", line 215, in get
bgr_fake = self.upscale_and_restore(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_swapping\upscaled_inswapper.py", line 153, in upscale_and_restore
upscaled = upscaling.upscale_img(pil_img, pp_options)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\extensions\sd-webui-faceswaplab\scripts\faceswaplab_postprocessing\upscaling.py", line 19, in upscale_img
result_image = pp_options.upscaler.scaler.upscale(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\upscaler.py", line 68, in upscale
img = self.do_upscale(img, selected_model)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\esrgan_model.py", line 36, in do_upscale
return esrgan_upscale(model, img)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\esrgan_model.py", line 57, in esrgan_upscale
return upscale_with_model(
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\upscaler_utils.py", line 74, in upscale_with_model
output = upscale_pil_patch(model, tile)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\modules\upscaler_utils.py", line 48, in upscale_pil_patch
return torch_bgr_to_pil_image(model(tensor))
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "C:\Users\mbord_\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\spandrel__helpers\model_descriptor.py", line 472, in call
output = self.call_fn(self.model, image)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\spandrel_helpers\model_descriptor.py", line 439, in
self.call_fn = call_fn or (lambda model, image: model(image))
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1532, in wrapped_call_impl
return self.call_impl(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1541, in call_impl
return forward_call(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\spandrel\architectures\ESRGAN\arch\RRDB.py", line 142, in forward
return self.model(x)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1532, in wrapped_call_impl
return self.call_impl(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1541, in call_impl
return forward_call(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\container.py", line 217, in forward
input = module(input)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1532, in wrapped_call_impl
return self.call_impl(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\module.py", line 1541, in call_impl
return forward_call(*args, **kwargs)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\extensions-builtin\Lora\networks.py", line 599, in network_Conv2d_forward
return originals.Conv2d_forward(self, input)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\conv.py", line 460, in forward
return self.conv_forward(input, self.weight, self.bias)
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\venv\lib\site-packages\torch\nn\modules\conv.py", line 456, in conv_forward
return F.conv2d(input, weight, bias, self.stride,
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\modules\dml\amp\autocast_mode.py", line 43, in
setattr(resolved_obj, func_path[-1], lambda *args, **kwargs: forward(op, args, kwargs))
File "C:\Users\mbord\Documents\stable-diffusion-webui-directml\modules\dml\amp\autocast_mode.py", line 15, in forward
return op(*args, **kwargs)
RuntimeError: Cannot set version_counter for inference tensor
To Reproduce
Venv built from scratch, clean install of A1111 and FaceSwap Lab.
Expected behavior
In inpaint tab of img2img, I usually set the attached options and it ran smoothly, producing swapped face. Now it's not possible to get an upscaled face.
Desktop (please complete the following information):
Ok, I think it's a problem of Directml - there has been an update recently.