independentopf.blogg.se

Change this block in lines 804-824 in visual_chatgpt.py ControlNet/cldm/ddim_hacked.py, line 20: attr = attr.to(vice("mps"))Īs well 3. ControlNet/annotator/midas/_init_.py:11: self.model = MiDaSInference(model_type="dpt_hybrid").to('mps') ControlNet/annotator/uniformer/_init_.py:18: self.model = init_segmentor(config_file, modelpath, 'mps') ControlNet/annotator/hed/_init_.py:103: Network = Network(modelpath).to('mps').eval()

Make the changes in these files (Thanks again )

Make sure you run the bash download.sh script in the readme before you patch the requirementsĪlso make sure to export your OPENAI_API_KEY and PYTORCH_ENABLE_MPS_FALLBACK=1 (temp fix for using stable diffusion on m1 macs with this pytorch imp see below) 2. I'm pretty sure some of these are not necessary since i see whisperx in there but idk how to clean it up remove package versions to allow pip attempt to solve the dependency conflictĮRROR: ResolutionImpossible: for help visit.loosen the range of package versions you've specified.Torchvision 0.14.1 depends on torch=1.13.1 Torchmetrics 0.11.3 depends on torch>=1.8.1 Pytorch-lightning 1.6.5 depends on torch>=1.8.* Invisible-watermark 0.1.5 depends on torch /ControlNet/cldm/ddim_hacked.py, line 20: attr = attr.to(vice("mps"))ĮRROR: Cannot install -r requirement.txt (line 1), -r requirement.txt (line 10), -r requirement.txt (line 11), -r requirement.txt (line 17), -r requirement.txt (line 19), -r requirement.txt (line 23), -r requirement.txt (line 24), -r requirement.txt (line 28), -r requirement.txt (line 29), -r requirement.txt (line 4) and torch=2.1.0.dev20230311 because these package versions have conflicting dependencies.image_depth = om_numpy(image_depth).float().to('mps')./ControlNet/annotator/midas/_init_.py:11: self.model = MiDaSInference(model_type="dpt_hybrid").to('mps')./ControlNet/annotator/uniformer/_init_.py:18: self.model = init_segmentor(config_file, modelpath, 'mps').and image_hed = om_numpy(input_image).float().to('mps')./ControlNet/annotator/hed/_init_.py:103: Network = Network(modelpath).to('mps').eval().(learned from this controlnet PR, chatgpt or perplexity were useless for this) I could publish a fork if people are interested but litreally follow the insturctions above thats all i didĮdit: welp another problem: image2line() calls MLSDdetector() which is cuda onlyĭiving into the source Controlnet/annotator/mlsd/_init_.py i think you have to change self.model = model.cuda().eval() to self.model = model.to('mps').eval().