Q-Learning Stock Trading Bot

This project demonstrates a simple Q-Learning algorithm applied to stock trading using Apple's historical price data. The agent learns to buy, sell, or hold shares to maximize profit over time.

What is Q-Learning?

Q-Learning is a reinforcement learning algorithm where an agent interacts with an environment, receives feedback, and learns the best actions to take. The agent updates its knowledge (Q-table) based on rewards received from its actions.

Below is a visual explanation of how Q-Learning works:

1. Input: The agent receives an input (state).
2. Response: The agent makes a prediction or takes an action.
3. Feedback: The environment provides feedback (reward or penalty).
4. Learns: The agent updates its knowledge based on feedback.
5. Reinforced Response: The agent improves its future responses.

How This Project Works

• Loads historical stock data for Apple.
• Sets up a trading environment with three actions: buy, sell, hold.
• Trains a Q-Learning agent to maximize profit.
• Tests the agent's performance on the data.

Usage

1. Place your dataset as dataset.csv in the project folder.

2. Run the script:

   python app.py

3. The script will display candlestick charts and print the trading results.

Requirements

• Python 3.x
• pandas
• numpy
• plotly

Install dependencies with:

pip install pandas numpy plotly

Files

• app.py: Main script with Q-Learning implementation.
• dataset.csv: Historical stock data.
• readme.md: Project documentation.

---