This project develops an advanced data mining pipeline tailored for stock trading, utilizing deep reinforcement learning algorithms including Deep Q-Network (DQN), Advantage Actor-Critic (A2C), and Proximal Policy Optimization (PPO). Our approach incorporates multiple reward functions and integrates sentiment analysis of financial news to enhance trading decisions. The project is currently a work in progress, and we are exploring innovative methods to integrate natural language processing with traditional numerical data analysis for stock market prediction.
google colab link : https://colab.research.google.com/drive/1W1GyJR_Rf9XeXdg0ektCOCksWRFmKm_9?usp=sharing,
Note: When executing the linked algorithms, it is important to be aware that multiple reward functions are available. To obtain results specific to each reward function, manually select and apply the desired function before running the algorithms. For final end result just run LLM_reward_testing.py
To set up the project environment and install the required dependencies, follow these steps:
# Clone the repository
git clone https://github.com/somsagar07/RL-stock-trading-.git
cd your-project-directory
# Install the required Python packages
pip install -r requirements.txt
Dataset Request: Since this is a work in progress we have not linked the dataset we used to train then RL agent with news yet. We plan to do this soon