# ytlabel **Repository Path**: minton95/ytlabel ## Basic Information - **Project Name**: ytlabel - **Description**: 宇图瑞视 YOLO标注训练工具，AI自动标注，大模型提示词自动标注，自动采集视频，自动收集数据集 - **Primary Language**: Python - **License**: MIT - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 166 - **Created**: 2026-10-03 - **Last Updated**: 2026-10-03 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # ytlabel **Language / 语言：** [简体中文](README.md) | [English](README_en.md) **License:** MIT License, free for commercial use. See `LICENSE` for details. - Website: https://www.yuturuishi.com - WeChat: yuturuishi - Gitee: https://gitee.com/yuturuishi/ytlabel - GitHub: https://github.com/yuturuishi/ytlabel - ytlabel is an open-source image annotation and model training tool built with Python + Flask, cross-platform on Windows / Linux / Mac. It supports multiple annotation types, AI-assisted auto-labeling, and the full YOLO training pipeline. --- ## Features - Annotation: multiple annotation types (rectangle, polygon, etc.); import images, videos, and LabelMe datasets - AI auto-labeling: large models (LMStudio, vLLM, ollama, Alibaba Cloud) for auto-labeling images and videos - Model training: full YOLO pipeline — dataset upload, training, resume-from-checkpoint, testing, parameter view and download - Datasets: export to YOLO format with a customizable train / val / test split - File management: built-in file manager with browse, upload, and download - Deployment: all static assets localized, supports offline deployment --- ## Screenshots 1 2 3 4 5 6 --- ## Requirements - Python 3.8+ - Dependencies in `requirements.txt` - Training requires Ultralytics / PyTorch (CPU or CUDA build) - A modern web browser --- ## Quick Start ```bash # Create and activate a virtual environment python -m venv venv # Windows venv\Scripts\activate # Linux / Mac source venv/bin/activate # Install base dependencies pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple ``` Start the server: ```bash python app.py --host 0.0.0.0 --port 9924 ``` Open `http://127.0.0.1:9924` in your browser. --- ## Usage 1. **Install dependencies**: see Quick Start. Training requires additional installs: ```bash pip install ultralytics==8.3.1 -i https://pypi.tuna.tsinghua.edu.cn/simple pip install numpy==1.26.4 -i https://pypi.tuna.tsinghua.edu.cn/simple # CPU build of torch pip install torch==2.1.2 torchvision==0.16.2 -i https://pypi.tuna.tsinghua.edu.cn/simple # CUDA build of torch pip install torch==2.1.0 torchaudio==2.1.0 torchvision==0.16.0 --index-url https://download.pytorch.org/whl/cu121 ``` 2. **Start server**: `python app.py --host 0.0.0.0 --port 9924` 3. **Open**: browser to http://127.0.0.1:9924 4. **Training**: visit http://127.0.0.1:9924/training — upload dataset → choose model → start training → test / download --- ## Project Structure ``` ytlabel/ ├── app.py # Main application ├── AiUtils.py # AI auto-labeling utility class ├── requirements.txt # Dependency list ├── static/ # Static assets (icons, styles, scripts, local Socket.IO) ├── templates/ # Page templates │ ├── index.html # Annotation home │ ├── training.html # Training panel │ ├── ai_config.html # AI configuration │ └── file_manager.html # File manager ├── pre_models/ # Pretrained models (.pt files) ├── uploads/ # Upload storage (created at runtime) │ ├── annotations/ # Annotation data │ ├── config/ # Config files │ ├── samples/ # Annotation images │ └── training_datasets/ # Training datasets ├── runs/ # Training output (created at runtime) └── tmp/ # Training temp files (created at runtime) ``` --- ## Shortcuts - **Ctrl+S**: save annotations - **Ctrl+Shift+D**: clear annotations --- ## Tech Stack Flask + Flask-SocketIO | HTML/CSS/JS | OpenCV/PIL | Ultralytics YOLO11 | Socket.IO --- ## Changelog Full changelog: [CHANGELOG.md](CHANGELOG.md) --- ## License The project's own code is released under the MIT License; keep the copyright notice to use it freely. Third-party libraries are subject to their respective licenses.