Skip to content

Latest commit

 

History

History
30 lines (21 loc) · 664 Bytes

README.md

File metadata and controls

30 lines (21 loc) · 664 Bytes

rnnTrader

rnnTrader should be an recurrent neural network cryptocurrency trader with predicting in the future.

Mostly based on https://pythonprogramming.net/cryptocurrency-recurrent-neural-network-deep-learning-python-tensorflow-keras/ but customized for the usage of cryptocurrency data from http://www.cryptodatadownload.com/index.html and more.

//TODO
Get live data from bitfinex
Backtesting
Trading simulation

Install

pip install -r requirements.txt

#To use bitfinex-api
pip install git+https://github.com/doghoc/bitfinex-api-py.git

Usage

python3 main.py

The trained models are located under {root}/models