Clean install of DLC following the new instructions for 3.0rc with PyTorch. Everything installs correctly, but when calling the GUI or importing DLC from python, it's still looking for a tensor flow ...
It pains me to see that people define a lazy loader for a module and then access it in the same module to load it immediately, but hiding that. I also had a hidden dependency on ml_dtypes somewhere ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results