Home

PyDeepFlow: A Lightweight Neural Network Framework

Welcome to PyDeepFlow, a lightweight and flexible Python framework for building deep learning models. This project is designed to provide users with the tools necessary to define and train custom Artificial Neural Networks (ANNs) with ease. Whether you're a beginner or an experienced ML practitioner, PyDeepFlow offers both CPU and GPU support, intuitive function abstractions, and comprehensive support for popular loss functions, activation functions, and learning rate schedules.

Author

Author Name: Ravin D
GitHub: ravin-d-27
Email: [email protected]

Contributors

Thanks to these amazing people for contributing to this project:

Key Features:

• Simplicity: Designed for ease of use, making it accessible to beginners.
• Easy-to-use API: Build custom multi-layer perceptrons with minimal code.
• Device Abstraction: Switch between CPU (NumPy) and GPU (CuPy) seamlessly.
• Customizable: Support for user-defined loss functions, activations, and learning rate schedules.
• Flexible Learning Rate Scheduling: Adjust learning rates during training using a variety of scheduling methods.

To learn more about how to use PyDeepFlow, explore the sections below:

1. Getting Started
2. Core Functions
3. Hyper Parameters
4. Examples
5. Advanced Features
6. Files Used in PyDeepFlow