This repository contains implementations of two advanced models for stock price prediction and automated trading:
- Price Prediction Model: Uses a transformer-based neural network to forecast future stock prices with high accuracy, utilizing technical indicators and historical price data.
- PPO-based Trading Environment: Employs a reinforcement learning agent (PPO) to make real-time trading decisions (Buy, Sell, Hold) in a simulated trading environment, aiming to maximize cumulative rewards based on market conditions.
Both models serve distinct yet complementary purposes, offering robust tools for financial analysis and automated trading strategies.
The dataset used for training the models is available at: XNAS ITCH Dataset
The model checkpoints are available on Hugging Face: Stock Recommendation with Transformer
You can run the models directly in Google Colab by clicking the button below:
