A centralized repository for production-ready Machine Learning projects and self-hosted utilities.
| Project | Description | Stack | Status |
|---|---|---|---|
| Fraud Detection | Imbalanced classification system for transaction security | Python, XGBoost, Streamlit | Production |
| Raw Voice | RAG-powered AI chatbot with voice processing | LangChain, GLM-4.7, ChromaDB | Production |
| Photonix | Self-hosted photo management system | Django, PostgreSQL, Redis | Production |
Ensure the following are installed on your system:
| Requirement | Purpose |
|---|---|
| Docker | Container runtime for all projects |
| Docker Compose | Multi-container orchestration |
| Python 3.12+ | Local development (optional) |
| Git LFS | Large dataset handling |
git clone https://github.com/Shreesh-Sree/ML_HUB.git
cd ML_HUBCreate a .env file in the root directory:
GOOGLE_API_KEY=your_google_api_key
HF_TOKEN=your_huggingface_token
LLM_MODEL=zai-org/GLM-4.7-Flash:novita
LLM_TEMPERATURE=0.7
RETRIEVER_K=5docker compose up -d| Service | URL | Description |
|---|---|---|
| Fraud Detection | localhost:8501 | Transaction fraud analysis dashboard |
| Raw Voice Frontend | localhost:8502 | AI chatbot interface |
| Raw Voice API | localhost:8000/docs | Backend API documentation |
Fraud Detection
cd credit-card-fraud-detection
docker build -t fraud-detection .
docker run -p 8501:8501 fraud-detectionAccess: http://localhost:8501
Raw Voice
cd raw_voice
docker compose up -dAccess: http://localhost:8502 (Frontend) | http://localhost:8000 (API)
ML_HUB/
├── credit-card-fraud-detection/ # XGBoost-based fraud classifier
├── raw_voice/ # LLM RAG chatbot with audio support
├── photonix/ # Photo management infrastructure
├── docker-compose.yml # Root orchestration file
└── .env # Environment configuration
This project is distributed under the MIT License. See individual project directories for specific licensing details.
Maintained by Sreesanth R