High Frequency Trading is a fundamental quantitative trading task, where traders actively buy/sell one pre-selected financial periodically in seconds with the consideration of order execution.
HFT_DDQN use a decayed supervised regulator genereated from the real q table based on the future price information and a double q network to optimizer the portfit margine.
In this notebook, we implement the training and testing process of HFTDDQN based on the TradeMaster framework.
And this is the script for training and testing.