Time Series Neural Networks (Keras wrapper)
# Time Series Neural Networks
[![Build Status](https://travis-ci.org/sofienealouini/tsnn.svg?branch=master)](https://travis-ci.org/sofienealouini/tsnn) [![Coverage Status](https://coveralls.io/repos/github/sofienealouini/tsnn/badge.svg?branch=master&service=github)](https://coveralls.io/github/sofienealouini/tsnn?branch=master)
TSNN is a deep learning library for time series forecasting built on Keras/Tensorflow. It implements various RNN-based models from recent research papers.
## Getting Started
The following instructions will get you a copy of the project up and running on your local machine.
Conda will set up a virtual environment with the exact version of Python used for development along with all the dependencies needed to run TSNN.
` conda create -n tsnn python=3.6 source activate tsnn `
Once you have activated your conda environment, you can easily install the package and all its dependencies from PyPI.
` pip install tsnn `
A comprehensive tutorial on how to use TSNN is provided PackageTesting.ipynb notebook.
## Built With
- [Keras](http://www.dropwizard.io/1.0.2/docs/) - High level Deep Learning library running on top of Tensorflow / Theano / CNTK
- [Tensorflow](https://maven.apache.org/) - Library for numerical computation, chosen as Keras backend in TSNN.
- Sofiene Alouini - Engineering graduate - Machine Learning Enthusiast