• Triton测试


    Trtion可以部署在cloud端,通过GRPC调用,也可以部署在边缘端,通过CAPI调用。

    1 启动docker

     这个以deepstreamtriton镜像为例。

    $ docker run --gpus all -itd --ipc=host --rm -v /tmp/.X11-unix:/tmp/.X11-unix  -v $HOME:/home   -e DISPLAY=$DISPLAY -w /opt/nvidia/deepstream/deepstream nvcr.io/nvidia/deepstream:6.1.1-triton

    2 GRPC方式

    2.1 启动tritonserver

    下载server的代码到本地,git clone https://github.com/triton-inference-server/server.git

    到docs目录下,下载模型,启动tritonserver.

    $ cd /home/code/triton/server/docs/examples

    $ ./fetch_models.sh

    $ tritonserver --model-repository=model_repository

    2.2 测试demo

    Releases · triton-inference-server/server · GitHub下载client sdk ,如: v2.25.0_ubuntu2004.clients.tar.gz,解压到本地,执行测试,

    $ ../bin/image_client -m inception_graphdef -s INCEPTION -c 3 -b 2 qa/images/mug.jpg

    返回结果如下:

    Request 0, batch size 2
    Image 'qa/images/mug.jpg':
        0.754047 (505) = COFFEE MUG
        0.157065 (969) = CUP
        0.002878 (968) = ESPRESSO
    Image 'qa/images/mug.jpg':
        0.754047 (505) = COFFEE MUG
        0.157065 (969) = CUP
        0.002878 (968) = ESPRESSO

    3 CAPI方式

    3.1 编译simple

    server目录的simple没有提供Makefile,按这个Makefile进行编译,缺少的东西需要添加。

    1. cat Makefile
    2. TARGET=simple
    3. GCC=g++
    4. GCC_PARMS+= -I./rdjson -I /usr/local/cuda-$(CUDA_VER)/include
    5. GCC_PARMS+=-I/opt/tritonserver/include -D TRITON_ENABLE_GPU=ON -D TRITON_MIN_COMPUTE_CAPABILITY=5.3
    6. GCC_LIBS= -L/usr/local/cuda-$(CUDA_VER)/lib64/ -L/opt/tritonserver/lib/
    7. GCC_LIBS+=-lpthread -ltritonserver -lcudart
    8. all: $(TARGET)
    9. %.o: %.cc
    10. $(GCC) $(GCC_PARMS) -c -g -o $@ $^
    11. $(TARGET): $(TARGET).o
    12. $(GCC) $^ $(GCC_LIBS) -o $@
    13. clean:
    14. rm -f $(TARGET).o $(TARGET)
    15. .PHONY: all clean

    3.2 测试simple

    编译成功后,传入模型路径给simple, 如: ./simple -r ../../docs/examples/model_repository/,部分结果打印如下:

    OUTPUT0 is stored in system memory
    OUTPUT1 is stored in system memory
    0 + 1 = 1
    0 - 1 = -1
    1 + 1 = 2
    1 - 1 = 0
    2 + 1 = 3
    2 - 1 = 1
    3 + 1 = 4
    3 - 1 = 2
    4 + 1 = 5
    4 - 1 = 3
    5 + 1 = 6
    5 - 1 = 4
    6 + 1 = 7
    6 - 1 = 5
    7 + 1 = 8
    7 - 1 = 6
    8 + 1 = 9
    8 - 1 = 7
    9 + 1 = 10
    9 - 1 = 8
    10 + 1 = 11
    10 - 1 = 9
    11 + 1 = 12
    11 - 1 = 10
    12 + 1 = 13
    12 - 1 = 11
    13 + 1 = 14
    13 - 1 = 12
    14 + 1 = 15
    14 - 1 = 13
    15 + 1 = 16
    15 - 1 = 14
     

  • 相关阅读:
    003.Vim编辑器
    ElasticSearch(超详细解说)[springBoot整合ES并简单实现增删改查]
    css中的z-index是什么
    企业如何保护机密文件安全
    438. 找到字符串中所有字母异位词
    计算机网络之IP数据格式(三)
    【CSS动效实战(纯CSS与JS动效)】03 精美手风琴侧边栏完整示例教程 示例1
    Ps:利用 AI 技术创建人像皮肤图层蒙版
    线程的概念+线程函数API
    【已解决】chrome视频无法自动播放的问题
  • 原文地址:https://blog.csdn.net/evsqiezi/article/details/127043811