type
status
date
slug
summary
tags
category
icon
password
LabelStudio部署
直接docker启动即可：
集成 Yolo
准备工作
- LabelStudio已启动
部署Label Studio ML backend
- 拉取仓库：
git clone <https://github.com/HumanSignal/label-studio-ml-backend.git>
- 修改
docker-compose.yml中以下变量 LABEL_STUDIO_URL: LabelStudio的启动地址，例如http://192.168.10.5:8080LABEL_STUDIO_API_KEY- 需要先在
Sidebar-Organization-API Token Settings中启动Legacy Tokens - 然后在账户设置中获取
- 启动docker服务：
docker compose up -d --build
LabelStudio中连接服务
- 项目设置里配置
<View> <Image name="image" value="$image"/> <RectangleLabels name="label" toName="image" model_path="yolov8n.pt" model_score_threshold="0.5"> <Label value="蜂鸟" predicted_values="bird" background="#FFA39E"/></RectangleLabels> </View>
- 连接模型服务：填写 LabelStudio 启动地址即可
- Author:Jarrett Au
- URL:https://blog.coding-park.top//article/labelstudio-deploy-guide
- Copyright:All articles in this blog, except for special statements, adopt BY-NC-SA agreement. Please indicate the source!
