A convolutional neural network (CNN) built from scratch using only NumPy to classify handwritten digits from the MNIST dataset.

An OS which is all about learning!

Implementation of KNN and Gaussian Naive-Bayes algorithms to classify phishing URLs. Built from scratch and compared with scikit-learn versions.

This project demonstrates how to build and train a feedforward neural network from scratch using only NumPy, without any high-level deep learning libraries like TensorFlow or PyTorch. The model is trained on the MNIST digit classification dataset and achieves competitive accuracy.

Minimal GPT implementation from scratch using PyTorch — trains a character-level transformer on the Tiny Shakespeare dataset to demonstrate core LLM concepts.

LSTM implemented from scratch and with PyTorch's nn.LSTM, trained using PyTorch Lightning on a toy stock prediction task. Educational and beginner-friendly.