search

city96 / ComfyUI-GGUF

GGUF Quantization support for native ComfyUI models
Apache License 2.0
1.08k stars 72 forks source link

incompatibility on python12 #158

Open michel-io opened 1 week ago

michel-io commented 1 week ago

running with --directml acceleration first got this error:

F:\ComfyUI\custom_nodes\ComfyUI-GGUF-main\nodes.py:79: UserWarning: The given NumPy array is not writable, and PyTorch does not support non-writable tensors. This means writing to this tensor will result in undefined behavior. You may want to copy the array to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\torch\csrc\utils\tensor_numpy.cpp:212.) torch_tensor = torch.from_numpy(tensor.data) # mmap

those i fix, but then an error where says that tensors are read only...

please fix

cchance27 commented 1 week ago

the numpy not writeable warning is suddenly showing for me as well on a clean install of comfy with gguf nodes, and im on a mac...

Python 3.11.5 pytorch version: 2.6.0.dev20241121

city96 commented 2 days ago

Is it possible that this is an issue specific to pytorch nightly (2.6.X) and not python 3.12? I did a quick test but at least on that specific env I can't reproduce it with torch 2.4.0