search

C0untFloyd / roop-unleashed

Evolved Fork of roop with Web Server and lots of additions
GNU Affero General Public License v3.0
2.13k stars 493 forks source link

Error init processor plugin dmdnet on CPU execution provider #91

Closed RiepinPavlo closed 1 year ago

RiepinPavlo commented 1 year ago

Describe the bug

Launching App Using provider ['CPUExecutionProvider'] - Device:cpu Running on local URL: http://127.0.0.1:7860

To create a public link, set share=True in launch(). Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: C:\Users\Admin/.insightface\models\buffalo_l\1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: C:\Users\Admin/.insightface\models\buffalo_l\2d106det.onnx landmark_2d_106 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: C:\Users\Admin/.insightface\models\buffalo_l\det_10g.onnx detection [1, 3, '?', '?'] 127.5 128.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: C:\Users\Admin/.insightface\models\buffalo_l\genderage.onnx genderage ['None', 3, 96, 96] 0.0 1.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: C:\Users\Admin/.insightface\models\buffalo_l\w600k_r50.onnx recognition ['None', 3, 112, 112] 127.5 127.5 set det-size: (640, 640) Error init processor plugin dmdnet... red Traceback (most recent call last): File "E:\nsfw-roop\roop-unleashed\installer\roop-unleashed\chain_img_processor\image.py", line 127, in init_processor self.add_processor_to_list(processor_id) File "E:\nsfw-roop\roop-unleashed\installer\roop-unleashed\chain_img_processor\image.py", line 113, in add_processor_to_list obj.init_plugin() File "E:\nsfw-roop\roop-unleashed\installer\roop-unleashed\plugins\plugin_dmdnet.py", line 72, in init_plugin create(self.device) File "E:\nsfw-roop\roop-unleashed\installer\roop-unleashed\plugins\plugin_dmdnet.py", line 179, in create weights = torch.load('./models/DMDNet.pth') File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 809, in load return _load(opened_zipfile, map_location, pickle_module, pickle_load_args) File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 1172, in _load result = unpickler.load() File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 1142, in persistent_load typed_storage = load_tensor(dtype, nbytes, key, _maybe_decode_ascii(location)) File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 1116, in load_tensor wrap_storage=restore_location(storage, location), File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 217, in default_restore_location result = fn(storage, location) File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 182, in _cuda_deserialize device = validate_cuda_device(location) File "E:\nsfw-roop\roop-unleashed\installer\installer_files\env\lib\site-packages\torch\serialization.py", line 166, in validate_cuda_device raise RuntimeError('Attempting to deserialize object on a CUDA ' RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU. To Reproduce** Steps to reproduce the behavior:

  1. Run python ./run with default options on CPU
  2. See error
  3. Start face swapping
  4. DMDnet not working/working strange - gives transparent grey square on face (Other post-process models works normal)

Details What OS are you using?

Are you try to use a GPU?

Sanity Check

eniora commented 1 year ago

I simply don't use DMDnet because of the transparent gray square alone (which you mentioned) and it tends to upscale the face in a way that looks so artificial, I know it's faster than others on CPU but I would rather not use it and instead use "None". On a side note for me GFPGAN is the best enhancer and much faster than codeformer.

C0untFloyd commented 1 year ago

Can't reproduce, working here with CPU too.