Skip to content

This program uses artificial recurrent neural netwrok called LongShort Term Memory (LSTM) to predict the closing price of BTCUSD pair from Bitstamp using 60 day past price

Notifications You must be signed in to change notification settings

LeoAlmDiniz/btcPriceNeuralNetwork

Repository files navigation

Case: BTC PRICE PREDICTION

This data analysis case was developed as result of the class ETL Fundamentals with Python from the bootcamp CARREFOUR BANK DATA ENGINEER from Digital Innovation One.

Description

In this project, BTC Price data from Quandl was queried to test its predictions for the dataset and for future data using LSTM Neural Network algorithm.

In order to run this Python code in your computer, install any version of Python 3 and pip install requirements.txt in your environment.

As this regression problem is non-convex, results vary according to the running environment. In this repository we present a example of graph obtained in one of these regressions (see priceHistory.png and 30dayPredict.png)

About

This program uses artificial recurrent neural netwrok called LongShort Term Memory (LSTM) to predict the closing price of BTCUSD pair from Bitstamp using 60 day past price

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages