ComfyUI
ComfyUI copied to clipboard
Dependency installation fails: No matching distribution found for torchdiffeq
OS: Arch Linux Rolling x86_64 GPU: AMD 6700XT Python: 3.11.3
(venv) ➜ ComfyUI git:(master) pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm5.5 -r requirements.txt
Looking in indexes: https://download.pytorch.org/whl/nightly/rocm5.5
Collecting torch
Downloading https://download.pytorch.org/whl/nightly/rocm5.5/torch-2.1.0.dev20230628%2Brocm5.5-cp311-cp311-linux_x86_64.whl (1420.3 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.4/1.4 GB 2.8 MB/s eta 0:00:00
Collecting torchvision
Downloading https://download.pytorch.org/whl/nightly/rocm5.5/torchvision-0.16.0.dev20230629%2Brocm5.5-cp311-cp311-linux_x86_64.whl (64.1 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 64.1/64.1 MB 4.4 MB/s eta 0:00:00
Collecting torchaudio
Downloading https://download.pytorch.org/whl/nightly/rocm5.5/torchaudio-2.1.0.dev20230629%2Brocm5.5-cp311-cp311-linux_x86_64.whl (2.6 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.6/2.6 MB 4.8 MB/s eta 0:00:00
ERROR: Could not find a version that satisfies the requirement torchdiffeq (from versions: none)
ERROR: No matching distribution found for torchdiffeq
WARNING: There was an error checking the latest version of pip.
Install pytorch with rocm5.5 then after do another pip install -r requirements.txt
Attempting install right now, my internet is a bit slow so it may take a bit. Unrelated to this issue, but I wonder if the Nvidia deps could be skipped if you don't have an Nvidia GPU, that'd save maybe half a gig?
Ok, so that just finished. I tried python main.py
but got this error:
(venv) ➜ ComfyUI git:(master) python main.py
Traceback (most recent call last):
File "/home/kainoa/.local/share/ComfyUI/main.py", line 28, in <module>
import execution
File "/home/kainoa/.local/share/ComfyUI/execution.py", line 12, in <module>
import nodes
File "/home/kainoa/.local/share/ComfyUI/nodes.py", line 19, in <module>
import comfy.diffusers_load
File "/home/kainoa/.local/share/ComfyUI/comfy/diffusers_load.py", line 6, in <module>
from comfy.sd import ModelPatcher, load_model_weights, CLIP, VAE, load_checkpoint
File "/home/kainoa/.local/share/ComfyUI/comfy/sd.py", line 6, in <module>
from comfy import model_management
File "/home/kainoa/.local/share/ComfyUI/comfy/model_management.py", line 103, in <module>
total_vram = get_total_memory(get_torch_device()) / (1024 * 1024)
^^^^^^^^^^^^^^^^^^
File "/home/kainoa/.local/share/ComfyUI/comfy/model_management.py", line 73, in get_torch_device
return torch.device(torch.cuda.current_device())
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/kainoa/.local/share/ComfyUI/venv/lib/python3.11/site-packages/torch/cuda/__init__.py", line 674, in current_device
_lazy_init()
File "/home/kainoa/.local/share/ComfyUI/venv/lib/python3.11/site-packages/torch/cuda/__init__.py", line 247, in _lazy_init
torch._C._cuda_init()
RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx
The commands I ran before were:
-
python3 -m venv venv
-
source ./venv/bin/activate
-
pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm5.5 -r requirements.txt
(this is what failed as per my first comment) -
pip install -r requirements.txt