Skip to content

Latest commit

 

History

History
29 lines (22 loc) · 1.1 KB

README.md

File metadata and controls

29 lines (22 loc) · 1.1 KB
Raw

Predicting Time Series using Neural Networks

🔵 Attn: For a curated list of Bitcoin learning materials, please follow this link and this other one

Introduction

Regardless of the speculative bubbles in Bitcoin price, this project aims to give a hint on the future closing price values of this cryptocurrency. Considering several Bitcoin indicators as per my analysis, sentiment data along with blockchain, historical prices and financial indices data are gathered to predict the closing price.

Environment Setup

To run the Prediction Models the following should be installed:

  1. Python 3+
  2. Tensorflow=1.10.1
  3. Keras=2.2.2
  4. Pandas=0.23.4
  5. Numpy=1.15.1
  6. Matplotlib=2.2.3
  7. sklearn=0.19.2

ML implemented Neural Network

LSTM