🏷️ What's ADE? (Auto Data Embedding) - In Progress

"Train. Detect. Automate."

A PyQt-based desktop app for automatic image labeling and embedding extraction using detection and vision models.

This tool helps you annotate image datasets quickly using pretrained models (e.g., YOLOv8~), modify labels with a GUI, and extract embeddings for downstream tasks.

Main View


Auto Embedding. . .


Train hyperparameter setting

📦 Features (planned)

✅ Automatic object detection & YOLO format label generation

✅ Visual label editing using PyQt GUI

✅ Embedding extraction (e.g., with CLIP or ResNet)

✅ Save embeddings as .npy, .json

✅ Support for .jpg, .png, .txt YOLO datasets

🔧 Setup Instructions

1. Download Release version

Release Date	Version
2025-06-12 - Recent	v1.0.2
2025-06-10	v1.0.1
2025-06-05	v1.0.0

2. Create and Activate a Virtual Environment

macOS/Linux

python3 -m venv venv
source venv/bin/activate

Windows

python -m venv venv
venv\Scripts\activate

3. Install Required Packages

pip install -r requirements.txt

Example requirements.txt

PyQt5==5.15.11
opencv-python==4.11.0.86
ultralytics==8.3.145
numpy==1.24.4
pillow==10.4.0
matplotlib==3.7.5

4. Install CUDA pytorch

pip install torch==2.4.1+cu121 torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu121

▶️ How to Build

pyinstaller main.py --onedir --noconsole --clean --icon=icons/icon.ico --add-binary "yolo.exe;."

📁 Project Structure

pyqt_app/
├── build_guide.txt               # Build guide
├── main.py                       # App entry point
├── data_integrator.py
├── training_settings_dialog.py
├── icons/                        # PyQt GUI icons
├── model/                        # YOLO, CLIP, etc. wrappers
├── dist/
│   └── main.exe                  # .exe file
├── dataset/
│   ├── images/
│   └── labels/
├── requirements.txt              # Requirements package settings
└── README.md

🧠 Model Support

• YOLOv11 via Ultralytics
• CLIP (planned)
• ResNet (planned)
• ONNX custom model support (planned)

🙌 Contributing

Issues, feedback, and pull requests are welcome! Open a GitHub issue or fork this repo to contribute.