Duarte Morais

Implemented various machine learning and deep learning algorithms, including perceptrons, logistic regression, multi-layer perceptrons, convolutional neural networks, and transformers, using both low-level libraries like NumPy and high-level frameworks such as PyTorch.

This project covers regression, showcasing the effectiveness of Orthogonal Matching Pursuit, and classification, addressing challenges in image classification with Convolutional Neural Networks and Support Vector Machines. Emphasizing solutions for imbalanced data, the project provides a thorough exploration of model performance.

A production-ready MLOps pipeline for stock price prediction using XGBoost. This project demonstrates the complete machine learning lifecycle, from data ingestion and feature engineering to model training, containerized deployment, and scalable inference serving on Kubernetes.