使用镜像起容器，然后git代码后，跑示例出错

[liuyiche]

1、 docker pull registry.baidubce.com/paddlepaddle/fastdeploy:1.0.4-gpu-cuda11.4-trt8.4-21.10 2、 docker run -it --net=host --name fastdeploy --gpus all --shm-size='10g' registry.baidubce.com/paddlepaddle/fastdeploy:1.0.4-gpu-cuda11.4-trt8.5-21.10 /bin/bash

3、 容器内git clone https://github.com/PaddlePaddle/FastDeploy.git

4、之后跑yolov8示例代码，

cpu报错如下： root@b385c568c32b:/FastDeploy/examples/vision/detection/yolov8/python# python3 infer.py --model yolov8.onnx --image 000000014439.jpg --device cpu [ERROR] fastdeploy/fastdeploy_model.cc(191)::CreateCpuBackend Found no valid backend for model: yolov8 [ERROR] fastdeploy/vision/detection/contrib/yolov8/yolov8.cc(40)::Initialize Failed to initialize fastdeploy backend. Traceback (most recent call last): File "infer.py", line 45, in model = fd.vision.detection.YOLOv8(args.model, runtime_option=runtime_option) File "/usr/local/lib/python3.8/dist-packages/fastdeploy/vision/detection/contrib/yolov8.py", line 181, in init assert self.initialized, "YOLOv8 initialize failed." AssertionError: YOLOv8 initialize failed.

gpu报错如下： root@b385c568c32b:/FastDeploy/examples/vision/detection/yolov8/python# python3 infer.py --model yolov8.onnx --image 000000014439.jpg --device gpu [ERROR] fastdeploy/fastdeploy_model.cc(214)::CreateGpuBackend Cannot find an available gpu backend to load this model. [ERROR] fastdeploy/vision/detection/contrib/yolov8/yolov8.cc(40)::Initialize Failed to initialize fastdeploy backend. Traceback (most recent call last): File "infer.py", line 45, in model = fd.vision.detection.YOLOv8(args.model, runtime_option=runtime_option) File "/usr/local/lib/python3.8/dist-packages/fastdeploy/vision/detection/contrib/yolov8.py", line 181, in init assert self.initialized, "YOLOv8 initialize failed." AssertionError: YOLOv8 initialize failed.

[DefTruth]

请问用的fastdeploy是安装的哪个版本呢？

[liuyiche]

请问用的fastdeploy是安装的哪个版本呢？

docker pull registry.baidubce.com/paddlepaddle/fastdeploy:1.0.4-gpu-cuda11.4-trt8.4-21.10 直接拉的这个镜像，1.0.4和1.0.3都试了

[liuyiche]

/FastDeploy/examples/vision/detection/yolov8/python# python3 infer.py --model yolov8.onnx --image 000000014439.jpg --device cpu

请问用的fastdeploy是安装的哪个版本呢？

示例代码就是按这个跑的： https://github.com/PaddlePaddle/FastDeploy/tree/develop/examples/vision/detection/yolov8/python

[fanruifeng]

解决了嘛 老哥 我现在问题也是和你一模一样

[fanruifeng]

是环境问题还是别的原因呢

[heliqi]

拉最新1.0.5的镜像还有问题吗？

[xzk-seu]

遇到一样的问题了，有没有好心人解答一下