paddle_serving_server 报错

[xiaomujiang]

在servering 启动服务的时候，出现了cuda找不到
PaddleDetection/serving_model#
export CUDA_VISIBLE_DEVICES=0
python -m paddle_serving_server_gpu.serve --model serving_model/ --thread 10 --port 9292 --gpu_ids 0
CUDA not found, please check your environment or use cpu version by "pip install paddle_serving_server"

版本如下
paddle-serving-app 0.1.1
paddle-serving-client 0.3.1
paddle-serving-server 0.3.1
paddle-serving-server-gpu 0.3.1
paddlehub 1.6.2
paddlepaddle-gpu 1.8.1.post107
python 3.7
查看显卡如下
nvidia-smi
Wed Jul 22 17:14:09 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.82 Driver Version: 440.82 CUDA Version: 10.2 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Quadro RTX 4000 Off | 00000000:5E:00.0 On | N/A |
| 30% 36C P8 2W / 125W | 603MiB / 7979MiB | 2% Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 27371 G /usr/lib/xorg/Xorg 169MiB |
| 0 28318 G /usr/bin/gnome-shell 199MiB |
| 0 33752 G ...AAAAAAAAAAAACAAAAAAAAAA= --shared-files 209MiB |
| 0 36758 G ...AAAAAAAAAAAACAAAAAAAAAA= --shared-files 22MiB |
+-----------------------------------------------------------------------------+

[barrierye]

您好 是在Serving提供的gpu镜像中启动server的吗？

[xiaomujiang]

您好 是在Serving提供的gpu镜像中启动server的吗？
我是参考这个https://aistudio.baidu.com/aistudio/projectdetail/643597做的，没有在GPU镜像中启动server， 是想将paddledetection产生的模型，用serving 进行启动， 但是没有启动起来，需要怎么做？

[barrierye]

可以根据这个文档导出Serving模型后，根据这个文档获取Docker镜像启动服务