dmbernaal / heisenberg

Stock Trading meets neural networks. This is the continuation of my previous work.

The notebooks are a bit of a mess but contain lots of detail someone may find userful. Otherwise, I will continue building on this repo to build a full-scale framework to train neural networks to predict stocks. Currently, I will be experimenting with transforming time-series data into images via GADF or GASF. Most recent experiments do show some promise but this requires extensive experimenting before going out there and going full diamondhands.

Disclaimer

This is for fun and should not be used to actually trade (unless you're a degen like myself).

Future Direction

Turning this into a full-blown framework to predict stock movement. This will contain models that run on technical indicators in different time windows and eventually using NLP for stock sentiment analysis.