Stock Price Forecasting Model

A machine learning project demonstrating end-to-end implementation of a deep learning pipeline for financial time-series forecasting. Built with PyTorch, this LSTM-based neural network uses 30-day historical sequences to predict next-day stock closing prices. The model was trained over 200 epochs using Mean Squared Error loss and Adam optimization on Apple stock data retrieved from Yahoo Finance's API.

The project showcases complete data science workflows including API integration for real-time market data retrieval, StandardScaler normalization for feature preprocessing, sequential data windowing for time-series modeling, and comparative visualizations of predicted versus actual price movements on held-out test data. Developed as a technical demonstration of Python's data science ecosystem including PyTorch, pandas, NumPy, scikit-learn, Matplotlib, and yfinance.