This project is an example of using PyTorch for stock quantitative analysis. It includes the following features:

1. Data acquisition: Obtain historical stock data with 5-minute granularity from www.alphavantage.co.

2. Data preprocessing: Construct technical indicator data from the raw stock data, including RSI, moving averages, KDJ, MACD, momentum indicators, volatility indicators, etc., and convert it to a format suitable for neural network models.

3. Model training: Train reinforcement learning-based neural network models using PyTorch.

4. Prediction and backtesting: Use the trained models for stock trading decision-making and backtesting.
(There is still much work to be accomplished.)

It's quite intriguing that 80% of the code was generated by ChatGPT.
Stupid humans + foolish robots = a bright future!!! Green: buy , Red: sell.