BPNN is a lightweight fully connected Backpropagation Neural Network implemented in NumPy, designed for regression tasks. It provides a simple, educational framework for experimenting with neural network fundamentals without relying on heavy deep learning libraries.
- Fully connected feedforward neural network
- Manual forward and backward propagation
- He initialization for weights
- Sigmoid activation for hidden layers, linear output for regression
- Standardization utilities to avoid data leakage
- Configurable number of layers and neurons
- Easy visualization of training loss and predictions
Clone the repository:
git clone https://github.com/lhrcom/bnpp.git
cd bpnnThis project is licensed under the MIT License. See LICENSE for details.
Contributions, issues, and feature requests are welcome! Feel free to fork the repo and submit pull requests.