GUGU-YT
commented
1 month ago it didn't work. :( respond same error RuntimeError: D:\a_work\1\s\onnxruntime\python\onnxruntime_pybind_state.cc:891 onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasnt able to be loaded. Please install the correct version of CUDA andcuDNN as mentioned in the GPU requirements page (https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html#requirements), make sure they're in the PATH, and that your GPU is supported.
I've tried to fix it for over 8h, but nothing worked. ONNX Runtime版本是1.18.1，CUDA版本是12.4，cuDNN版本是9.3.0，torch版本是2.4.0，torchaudio版本是2.4.0，torchvision版本是0.19.0，显卡驱动版本是555.85，为什么使用SD的WD 14 标签 这个插件会显示onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasnt able to be loaded. 版本→version
C:\Users\Administrator>echo %CUDA_PATH% C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4
import torch print(torch.cuda.is_available()) True
CUDA_PATH C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4
CUDA_PATH_V12_4 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4
import onnxruntime as ort providers = ort.get_available_providers() print(providers) ['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']
"I've checked everything thoroughly, following the official documentation, but I still can't use the reverse prompt suggestion plugin. Interestingly, this plugin was working before I installed CUDA. I'm wondering if the issue could be with the plugin itself?"
I also tried: https://github.com/toshiaki1729/dataset-tag-editor-standalone https://github.com/67372a/stable-diffusion-webui-wd14-tagger They respond the same error.
Maybe I can create a new environment using Conda that doesn't call CUDA? I'm not sure if this approach will work since this is my first time using Conda; before, I always ran everything directly on the local machine.
把cuda版本降到11.8即可解决，但是还没试能不能加载依赖。
slashedstar
Cheesper
it didn't work. :( respond same error RuntimeError: D:\a_work\1\s\onnxruntime\python\onnxruntime_pybind_state.cc:891 onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasnt able to be loaded. Please install the correct version of CUDA andcuDNN as mentioned in the GPU requirements page (https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html#requirements), make sure they're in the PATH, and that your GPU is supported.
I've tried to fix it for over 8h, but nothing worked. ONNX Runtime版本是1.18.1，CUDA版本是12.4，cuDNN版本是9.3.0，torch版本是2.4.0，torchaudio版本是2.4.0，torchvision版本是0.19.0，显卡驱动版本是555.85，为什么使用SD的WD 14 标签 这个插件会显示onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasnt able to be loaded. 版本→version
C:\Users\Administrator>echo %CUDA_PATH% C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4
import torch print(torch.cuda.is_available()) True
CUDA_PATH C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4
CUDA_PATH_V12_4 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4
import onnxruntime as ort providers = ort.get_available_providers() print(providers) ['TensorrtExecutionProvider', 'CUDAExecutionProvider', 'CPUExecutionProvider']
"I've checked everything thoroughly, following the official documentation, but I still can't use the reverse prompt suggestion plugin. Interestingly, this plugin was working before I installed CUDA. I'm wondering if the issue could be with the plugin itself?"
I also tried: https://github.com/toshiaki1729/dataset-tag-editor-standalone https://github.com/67372a/stable-diffusion-webui-wd14-tagger They respond the same error.
Maybe I can create a new environment using Conda that doesn't call CUDA? I'm not sure if this approach will work since this is my first time using Conda; before, I always ran everything directly on the local machine.