Uninet is an open-source project that aims to unify a wide range of machine learning tasks and neural network architectures under a single, extensible framework. Whether you're solving problems in classification, regression, clustering, or physics-based modeling, Uninet provides modular components and pre-built templates to accelerate your development.
- ✅ Task-Oriented Modules: Easily switch between classification, regression, clustering, and more.
- 🧩 Pluggable Network Architectures: Use or extend support for CNNs, RNNs, Transformers, GANs, PINNs, GNNs, and more.
- 🔬 Physics-Informed Learning: Solve differential equations using integrated PINNs support.
- 📈 Unified Training & Evaluation Interface: Consistent APIs across different models and datasets.
- 🧠 Model Zoo: Predefined models for rapid prototyping.
- 🔁 AutoML-ready: (Planned) Integration with hyperparameter tuning frameworks.
- 📚 Educational Mode: Interactive notebooks to learn concepts as you build.
Complete platform functionality demonstration including ML Tasks, Various Neural Neural Networks, and Example Training
- Feedforward Neural Networks (FNN)
- Convolutional Neural Networks (CNN)
- Recurrent Neural Networks (RNN, LSTM, GRU)
- Transformer Models
- Generative Adversarial Networks (GAN)
- Autoencoders & VAEs
- Graph Neural Networks (GNN)
- Physics-Informed Neural Networks (PINN)
- Neural ODEs
- More coming soon...
- Classification
- Regression
- Clustering
- Dimensionality Reduction
- Sequence Modeling
- Anomaly Detection
- Generative Modeling
- Reinforcement Learning (Planned)
- Physics-Based Modeling
uninet/
├── models/ # Modular neural network architectures
├── tasks/ # Task-specific pipelines (classification, etc.)
├── datasets/ # Preprocessing and loading
├── utils/ # Shared utilities
├── notebooks/ # Example tutorials and walkthroughs
├── configs/ # Training and model config files
├── README.md
└── setup.py
# Clone the repository
git clone https://github.com/your-username/uninet.git
cd uninet
# Set up environment
pip install -r requirements.txt
# Run a sample classification task
python tasks/classification/train.py --config configs/mnist_cnn.yaml- CNN on MNIST (Classification)
- RNN for Time-Series Forecasting
- Transformer for Text Classification
- GAN for Image Generation
- PINN for Solving a Heat Equation
(More coming soon...)
Uninet is designed to be modular and community-driven. Contributions are welcome — whether it's bug fixes, new models, or entire task modules!
Would you like a README.md file auto-generated or customized badges (e.g., license, build, Python version) for GitHub?