DynamicForexTrader is an automated trading system designed to operate on the Foreign Exchange Market (FOREX) through the utilization of Python, effectively interfacing with the OANDA API and tpqoa for trading execution. It features a class-oriented strategy implementation, empowering traders to freely create an array of distinct strategies, with the flexibility to specify/adjust specific parameters to dynamically mutate the operational trading strategy.
The system also includes a binary classification deep neural network (DNN). I engineered and trained the DNN on over 40,000 data points, which utilizes binary cross-entropy loss and sigmoid activation for market prediction with TensorFlow, Keras, and scikit-learn.
If you decide to clone the repository, please run:
pip install -r requirements.txt to download the required libraries for the project.
Also, you will need to create an Oanda configuration file, named oanda.cfg which is filled with your account and API key information. It should be structured as follows:
[oanda]
account_id = XXX-XXX-XXXXXXXX-XXX
access_token = XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
account_type = practice (default) or live