C# code for cAlgo / cTrader Automated Trading Platform
Proof of concept to use neural networks to predict profitable market moves implemented for the cTrader/cAlgo platform. A simple feed forward NN was trained on features that are based on prices, to predict market moves above a threshold in the NNv3.py file. The trained network saved as xml in file Trained_NNv4.xml The cTrader / cAlgo code NN_call_v1.cs, calls the python script ReadAndUseNNv4.py with the latest price features, which in turn returns a prediction on if and what direction to trade. The trade is executed on the cTrader platform.
for write up see http://www.peoplesfintech.com/blog/2019/05/02/using-neural-networks-in-trade-recommendation-systems-a-proof-of-concept-study-%EF%BB%BF/