Skip to content
David edited this page Sep 8, 2020 · 1 revision

Welcome to the Tensorflow-LSTM-Bitcoin-forecast wiki!

How it works

We gather financial information from yahoo finance (thank you to the yfinance library and general information such as historical data of google search (thank you for the keyword "Bitcoin" and turn it into a Panda Dataframe.

Then we define a dataset based on the Dataframe using a Class called "WindowGenerator"

Last, an LSTM neural network is trained on the dataset to predict the next day's Bitcoin value (should not be taken for granted for trading!)

The system is based on Tensorflow 2.0

Clone this wiki locally