This project is a companion to a blog series on Machine Learning and AI foundations. It builds knowledge from the ground up on how neural networks and AI (Large Language Models in particular) work.
We build knowledge progressively, starting with Deep Neural Networks, Recurrent Neural Networks, and lastly Large Language Models
To familiarize yourself with various concepts, each topic is approached from first principles. Rather than jumping straight to implementations using PyTorch or TensorFlow, we start with vanilla implementations.
For example, neural network models can be trained using implementations from different approaches:
- Vanilla Python (standard library only)
- Python + NumPy
- PyTorch
- TensorFlow
See each package directory for instructions on running the various implementations.
This is an open-source project and contributions are welcome.
Bug Fixes: Please create a pull request explaining the bug, how you fixed it, and how you tested it.
New Features: Please open an issue with the enhancement label, detailing the feature and explaining why it's needed. We'll discuss it there, and once approved, you can create a pull request.
Chido Warambwa - Initial work - chidow@centridsol.tech
This project is licensed under the MIT License - see the LICENSE file for details.