Stock price forecasting project using Time Series Foundation Models (Chronos, TimeFM, Uni2TS), with performance comparison against benchmark models.

A stock price forecasting system using a custom Transformer model built with PyTorch. Predicts next-day returns and close prices for GOOG, TSLA, and SPY using technical indicators as input features.

Hybrid AI crypto price forecasting system using Transformer and LSTM models with an interactive Streamlit dashboard for probabilistic predictions and financial time-series visualization.