【AI换脸】一键换脸 FaceFusion2.6.0 N卡版 本地一键整合包下载

w41***

dfsgsfdgsfdgsdfg

weihu1***

请问为啥刚下载了你发的版本，选择的是cuda为啥是cpu在跑呢

weihu1***

111111试试

Ar***

weihu19940327 发表于 2024-8-19 11:59
请问为啥刚下载了你发的版本，选择的是cuda为啥是cpu在跑呢

我也碰到这个问题，安装了cuda和cudnn后，就再也运行不了

出错提示：

Running on local URL:  http://127.0.0.1:7860
2024-09-10 00:09:43.1035172 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

EP Error D:\a\_work\1\s\onnxruntime\python\onnxruntime_pybind_state.cc:743 onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasn't able to be loaded. Please install the correct version of CUDA and cuDNN 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.
when using [('CUDAExecutionProvider', {'device_id': '0', 'cudnn_conv_algo_search': 'DEFAULT'})]
Falling back to ['CUDAExecutionProvider', 'CPUExecutionProvider'] and retrying.
2024-09-10 00:09:43.2558304 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

Traceback (most recent call last):
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 419, in __init__
    self._create_inference_session(providers, provider_options, disabled_optimizers)
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 463, in _create_inference_session
    sess.initialize_session(providers, provider_options, disabled_optimizers)
RuntimeError: D:\a\_work\1\s\onnxruntime\python\onnxruntime_pybind_state.cc:743 onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasn't able to be loaded. Please install the correct version of CUDA and cuDNN 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.


The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\gradio\queueing.py", line 407, in call_prediction
    output = await route_utils.call_process_api(
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\gradio\route_utils.py", line 226, in call_process_api
    output = await app.get_blocks().process_api(
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\gradio\blocks.py", line 1550, in process_api
    result = await self.call_function(
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\gradio\blocks.py", line 1185, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\gradio\utils.py", line 661, in wrapper
    response = f(*args, **kwargs)
  File "C:\FaceFusion_2.6.0\facefusion\uis\components\preview.py", line 178, in update_preview_image
    preview_vision_frame = process_preview_frame(reference_faces, source_face, source_audio_frame, temp_vision_frame)
  File "C:\FaceFusion_2.6.0\facefusion\uis\components\preview.py", line 200, in process_preview_frame
    target_vision_frame = frame_processor_module.process_frame(
  File "C:\FaceFusion_2.6.0\facefusion\processors\frame\modules\face_enhancer.py", line 261, in process_frame
    target_vision_frame = enhance_face(target_face, target_vision_frame)
  File "C:\FaceFusion_2.6.0\facefusion\processors\frame\modules\face_enhancer.py", line 204, in enhance_face
    crop_vision_frame = apply_enhance(crop_vision_frame)
  File "C:\FaceFusion_2.6.0\facefusion\processors\frame\modules\face_enhancer.py", line 213, in apply_enhance
    frame_processor = get_frame_processor()
  File "C:\FaceFusion_2.6.0\facefusion\processors\frame\modules\face_enhancer.py", line 107, in get_frame_processor
    FRAME_PROCESSOR = onnxruntime.InferenceSession(model_path, providers = apply_execution_provider_options(facefusion.globals.execution_device_id, facefusion.globals.execution_providers))
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 430, in __init__
    raise fallback_error from e
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 425, in __init__
    self._create_inference_session(self._fallback_providers, None)
  File "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 463, in _create_inference_session
    sess.initialize_session(providers, provider_options, disabled_optimizers)
RuntimeError: D:\a\_work\1\s\onnxruntime\python\onnxruntime_pybind_state.cc:743 onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasn't able to be loaded. Please install the correct version of CUDA and cuDNN 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.

Ar***

把环境变量CUDA_PATH 注释掉后，又可以用，但是选CUDA，CPU用100%，GPU只用了1-2%，而且非常慢

Ar***

Arwar 发表于 2024-9-10 00:51
把环境变量CUDA_PATH 注释掉后，又可以用，但是选CUDA，CPU用100%，GPU只用了1-2%，而且非常慢
...

把环境参数注释掉后的运行代码，CUDA也是用不了的

-----------------------------------------------
  更多好玩的AI应用，访问 https://deepface.cc
  FaceFusion WebUI正在启动中，请稍后
-----------------------------------------------
Running on local URL:  http://127.0.0.1:7860
2024-09-10 00:47:46.5897328 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

2024-09-10 00:47:57.8229684 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

正在分析: 100%|========================================================| 2409/2409 [00:11<00:00, 201.97frame/s, rate=0]
[FACEFUSION.CORE] 正在提取分辨率为720x1280，每秒25.0帧的帧
[FACEFUSION.PROCESSORS.FRAME.MODULES.FACE_ENHANCER] 正在处理
正在处理:   0%| | 0/2409 [00:00<?, ?frame/s, execution_providers=['cuda'], execution_thread_count=4, execution_queue_co2024-09-10 00:48:33.9558129 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

2024-09-10 00:48:34.2011241 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

2024-09-10 00:48:34.5614814 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

2024-09-10 00:48:35.1830855 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

2024-09-10 00:48:35.2472554 [Ennxruntimeefault, provider_bridge_ort.cc:1480 onnxruntime::TryGetProviderInfo_CUDA] D:\a\_work\1\s\onnxruntime\core\session\provider_bridge_ort.cc:1193 onnxruntime:roviderLibrary::Get [ONNXRuntimeError] : 1 : FAIL : LoadLibrary failed with error 126 "" when trying to load "C:\FaceFusion_2.6.0\python310\lib\site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"

正在处理:   8%| | 188/2409 [05:41<39:33,  1.07s/frame, execution_providers=['cuda'], execution_thread_count=4, executio

无言***

Arwar 发表于 2024-9-10 00:56
把环境参数注释掉后的运行代码，CUDA也是用不了的

-----------------------------------------------

因为目前的主流AI项目都是基于N卡的 CUDA 生态，所以，接下来就要安装 CUDA (CUDA非必要不要安装12.X，很多AI程序不兼容) 和 cuDNN，这样就能确保一键包的正常运行。


windows下CUDA与cuDNN和安装和配置：https://deepface.cc/thread-268-1-1.html  视频教程 https://www.bilibili.com/video/BV1wm41167w8/

Ar***

确是按指引安装的成功了，运行时就出错，请问是cuda版本的原因吗？

无言***

Arwar 发表于 2024-9-10 09:35
确是按指引安装的成功了，运行时就出错，请问是cuda版本的原因吗？

报错就是cuda没安装
不知道是不是12.4版本高了的问题，你可以试试再装个11.8，把11.8设置为默认
https://blog.csdn.net/weixin_42672685/article/details/138814153

Ar***

应该是版本兼容的问题，可以确定是成功安装了的，运行另一个MuseTalk模型，比没安装cuda前快了很多，GPU拉满。


运行nvcc --version
C:\Users\MSI-NB>nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Thu_Mar_28_02:30:10_Pacific_Daylight_Time_2024
Cuda compilation tools, release 12.4, V12.4.131
Build cuda_12.4.r12.4/compiler.34097967_0