from-scratch-ml

Companion code for Machine Learning From Scratch — 10 core ML algorithms built from scratch with NumPy, compared with Scikit-learn and PyTorch.

Academic implementation of a Multi-Layer Perceptron from scratch using Python and NumPy.

My first artificial neuron built from scratch in Python. It learns to approximate a linear function from data points using gradient descent, MSE, and manually calculated gradients, with a real-time visualization of the learning process.

CUDA accelerated doodle recognition neural network built from scratch in C++/CUDA with custom model inference.

End-to-end ML pipeline for California house price prediction. Features engineered data, OLS/Ridge/Lasso models, custom Gradient Descent, cross-validation, hyperparameter tuning, and a full training workflow.

A hands-on implementation of Linear and Polynomial Regression from scratch using the real-world California Housing dataset. Includes comparisons of various optimization algorithms and professional libraries like Scikit-Learn and PyTorch.

Predicting clinical trial duration from registration data. Neural network built from scratch in NumPy, benchmarked against sklearn and Keras.

🪐 Return players safely from the End to the Overworld when they fall or teleport, ensuring a smooth transition with simple, configurable options.

A simple neural network built from scratch in Python to learn the XOR function. It implements forward propagation, backpropagation, sigmoid activation, gradient descent, and training without using machine learning libraries.