requirements changes torch to cpu style .can't run comfy ui
cnhh-cyc opened this issue · 2 comments
before \python_embeded\python.exe -m pip install -r requirements.txt
import torch
print(torch.version)
2.2.0+cu121
torch.cuda.is_available()
True
安装依赖之后
after install requirements.txt
print(torch.version)
2.3.0+cpu
torch.cuda.is_available()
False
and can't run comfy ui:
File "D:.....\new_ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda_init_.py", line 284, in _lazy_init
raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled
彻底打不开comfy ui了,
same
i have got it
我直接用中文吧,
因为需要安装torch的版本是2.3.0,所以自动给升级装了个2.3.0的cpu版本
需要去\new_ComfyUI_windows_portable\python_embeded\Scripts 下面cmd,
然后用python -m pip show torch或者pip show torch查看
理论是用了python -m是局部python环境,不加的是系统环境。但是我这台不知道怎么回事,反了一下,那就是看哪个命令下出现的torch的location是在new_ComfyUI_windows_portable\python_embeded\下还是系统下
比如可能会出现
Location: 。。。。。\appdata\roaming\python\python310\site-packages
或者
。。。。。new_ComfyUI_windows_portable\python_embeded\Scripts>pip show torch
Name: torch
Version: 2.3.0
Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration
Home-page: https://pytorch.org/
Author: PyTorch Team
Author-email: packages@pytorch.org
License: BSD-3
Location: 。。。。。\new_ComfyUI_windows_portable\python_embeded\Lib\site-packages
显然是下面那个是对的,那就用相应的命令去安装
(
以此判断到底要不要用python -m
python -m pip install torch-2.2.0+cu121-cp310-cp310-win_amd64.whl
或者直接
pip install torch-2.2.0+cu121-cp310-cp310-win_amd64.whl
)否则你会装了个寂寞,应该装在局部python环境下结果装到了全局环境下
注意1局部python环境的版本,comfyui应该是3.11,我之前预下载python 3.10的torch,结果显然装不上,报错一次
注意2 必须2.3.0的torch,否则comfyui可以运行,但是动画是生不成的,因为后面依赖的是2.3.0的torch
torch地址在下面:https://pytorch.org/get-started/previous-versions/
ROCM 6.0 (Linux only)
pip install torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/rocm6.0
CUDA 11.8
pip install torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/cu118
CUDA 12.1
pip install torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/cu121
CPU only
pip install torch==2.3.0 torchvision==0.18.0 torchaudio==2.3.0 --index-url https://download.pytorch.org/whl/cpu