eseca/LSTM-Neural-Network-for-Time-Series-Prediction

LSTM built using Keras Python package to predict time series steps and sequences. Includes sin wave and stock market data

LSTM Neural Network for Time Series Prediction

LSTM built using the Keras Python package to predict time series steps and sequences. Includes sin wave and stock market data.

Full article write-up for this code

Video on the workings and usage of LSTMs and run-through of this code

Requirements

Update: As of 08-03-2017 this code has been updated from Python 2.7 to Python 3.5.

This is to make it available to run natively on Windows TensorFlow installations (Tutorial on how to install TensorFlow natively on Windows)

Installation

Install virtualenv and create a new virtual environment.

$ virtualenv venv

Activate de virtual environment.

$ source venv/bin/activate

Install de requirements.

$ pip install -r requirements.txt

Run

To run the example:

$ python run.py

Output for sin wave sequential prediction:

Output for stock market multiple sequential predictions: