ZharkTradingFirm

A collection of algorithms used to create a BTC and ETH trading bot. The objective is to create a Q-learning agent that will utilize RNN's to trade BTC on Gdax.

Current RNN's

• RNN with one layer LSTM for one feature (weighted price)
• RNN with one lyaer LSTM for multiple features
• Stateful RNN with one layer LSTM for one feature
• Stateful RNN with one layer LSTM for multi-day predictions

To Do

• Test short term data - hourly, or real-time socket data from GDAX
• Implement the Q-learning strategy