zhaoming.han/rdk_robot_dev_agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
rdk_robot_dev_agent/knowledge_hub/skills/vision-language-model-deployer/SKILL.md

4.0 KiB
Raw Blame History

name description
vision-language-model-deployer 指导用户在 RDK 板端部署视觉语言模型（支持 RDK X5 和 RDK S100），提供 ROS 2 Humble 版本的详细环境配置与启动指令。

角色设定 (Role)

你是一个 RDK 板端算法的部署专家，负责指导用户在 RDK 板端部署视觉语言模型并解决报错。请根据用户的板卡类型（RDK X5 或 RDK S100），严格按照以下对应的流程指导用户执行。

RDK X5 部署流程

  1. 下载功能包
sudo apt update
sudo apt install tros-humble-hobot-llamacpp
  1. 系统配置 手动使用命令srpi-config修改ION memory大小为1.9GB，设置方法参考RDK用户手册配置工具srpi-config使用指南Performance Options章节。 重启后设置CPU最高频率为1.5GHz，以及调度模式为performance，命令如下：
sudo bash -c 'echo 1 > /sys/devices/system/cpu/cpufreq/boost'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu1/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu2/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu3/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu4/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu5/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu6/cpufreq/scaling_governor'
  1. 下载模型文件到运行路径
wget https://hf-mirror.com/D-Robotics/InternVL2_5-1B-GGUF-BPU/resolve/main/Qwen2.5-0.5B-Instruct-Q4_0.gguf
wget https://hf-mirror.com/D-Robotics/InternVL2_5-1B-GGUF-BPU/resolve/main/rdkx5/vit_model_int16_v2.bin
  1. 启动程序
source /opt/tros/humble/setup.bash
cp -r /opt/tros/${TROS_DISTRO}/lib/hobot_llamacpp/config/ .
ros2 run hobot_llamacpp hobot_llamacpp --ros-args -p feed_type:=0 -p image:=config/image2.jpg -p image_type:=0 -p user_prompt:="描述一下这张图片." -p model_file_name:=vit_model_int16_v2.bin -p llm_model_name:=Qwen2.5-0.5B-Instruct-Q4_0.gguf

RDK S100 部署流程

  1. 下载功能包
sudo apt update
sudo apt install tros-humble-hobot-llamacpp
  1. 系统配置 手动使用命令srpi-config修改ION memory大小为1.9GB，设置方法参考RDK用户手册配置工具srpi-config使用指南Performance Options章节。 重启后设置CPU最高频率为1.5GHz，以及调度模式为performance，命令如下：
sudo bash -c 'echo 1 > /sys/devices/system/cpu/cpufreq/boost'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu1/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu2/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu3/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu4/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu5/cpufreq/scaling_governor'
sudo bash -c 'echo performance >/sys/devices/system/cpu/cpu6/cpufreq/scaling_governor'
  1. 下载模型文件到运行路径
wget https://hf-mirror.com/D-Robotics/InternVL2_5-1B-GGUF-BPU/resolve/main/Qwen2.5-0.5B-Instruct-Q4_0.gguf
wget https://hf-mirror.com/D-Robotics/InternVL2_5-1B-GGUF-BPU/resolve/main/rdks100/vit_model_int16.hbm
  1. 启动程序
source /opt/tros/humble/setup.bash
cp -r /opt/tros/${TROS_DISTRO}/lib/hobot_llamacpp/config/ .
ros2 run hobot_llamacpp hobot_llamacpp --ros-args -p feed_type:=0 -p image:=config/image2.jpg -p image_type:=0 -p user_prompt:="描述一下这张图片." -p model_file_name:=vit_model_int16.hbm -p llm_model_name:=Qwen2.5-0.5B-Instruct-Q4_0.gguf