GrowEngine 🚀

中文文档

GrowEngine is a comprehensive advertising automation platform evolved to include a Hybrid Recommendation System. It combines a robust ad campaign management dashboard with an experimental, high-performance recommendation engine integrated from SparrowRecSys.

✨ Features

🏢 Core Platform

• 📊 Real-time Dashboard: Monitor Spend, GMV, ROI, and core metrics instantly.
• 📋 Campaign Management: Full lifecycle management for ad plans (Create, Edit, Pause/Resume).
• 🤖 Smart Diagnosis: AI-driven optimization suggestions for ad performance.
• 🎮 Bidding Simulation: Simulate bidding strategies based on OnlineLp algorithms.

🎯 Hybrid Recommendation System (New)

Integrated directly with SparrowRecSys, offering a "Hybrid Mode" architecture:

• Python Backend: Handles Data Management, Feature Engineering, and Click Collection.
• Model Inference: Supports Java RecSysServer or TensorFlow Serving for NeuralCF, DeepFM, and DIN models.
• Interactive UI: Dedicated React frontend for visualizing recommendations and tracking user interactions.

🏗 Architecture

GrowEngine now operates on a modular architecture:

graph TD
    User["User/Visitor"]
    
    subgraph "Ad Management"
        AdminUI["Admin Dashboard"]
        CoreAPI["Core API (8000)"]
    end
    
    subgraph "Recommendation Engine"
        RecUI["RecSys UI"]
        RecAPI["RecSys API (8001)"]
        JavaServ["Java RecSysServer (6010)"]
    end
    
    User --> AdminUI
    User --> RecUI
    
    AdminUI --> CoreAPI
    RecUI --> RecAPI
    
    RecAPI --> JavaServ
    RecAPI -- "Click Logs" --> Data[("CSV/Clicks")]
    Data -- "Training" --> Train[Model Pipeline]
    Train --> JavaServ

Loading

📁 Project Structure

ProtoAd/
├── frontend/          # Admin Dashboard (React + Vite + TailwindCSS)
├── backend/           # Core Ad Management API (FastAPI)
├── ad_rec_frontend/   # Recommendation System Showcase UI (React)
├── ad_rec_backend/    # RecSys API, Click Collection & Data Mgr
├── SparrowRecSys/     # Original Java Recommendation Engine Source
├── web_visualization/ # Helper visualizations
├── scripts/           # DevOps and utility scripts
└── docker-compose.yml # Full stack orchestration

🚀 Getting Started

Prerequisites

• Node.js 18+
• Python 3.10+
• Docker (Optional, for full stack deployment)

💻 Local Development

1. Core Platform (Management)

Backend:

cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python generate_mock_data.py
uvicorn api:app --reload --port 8000

Frontend:

cd frontend
npm install
npm run dev
# Access at http://localhost:5173

2. Recommendation System Module

RecSys Backend:

cd ad_rec_backend
# Ensure dependencies are installed (or reuse backend venv if compatible)
uvicorn api:app --reload --port 8001

RecSys Frontend:

cd ad_rec_frontend
npm install
npm run dev
# Access at http://localhost:3000 (check console for actual port)

🐳 Docker Deployment

Run the entire suite with one command:

docker-compose up -d --build

This will start:

• Core Backend
• Core Frontend
• (Configuration dependent) RecSys services

📡 API Documentation

Service	Base URL	Documentation
Core API	http://localhost:8000	/docs
RecSys API	http://localhost:8001	/docs

Key RecSys Endpoints

• GET /api/rec/ads: Get personalized ad recommendations.
• GET /api/rec/similar: Get similar items (Item2Vec).
• POST /api/rec/click: Log user interaction events.
• POST /api/rec/train: Trigger model retraining pipeline.

📈 Star History

🤝 Contributing

We welcome contributions!

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📄 License

Distributed under the MIT License. See LICENSE for more information.