Temporal convolutional networks applied to energy-related time series forecasting

Authors: @pedrolarben and @carranza96

This repository contains the source code from the following paper.

Lara-Benítez, P.; Carranza-García, M.; Luna-Romera, J.M.; Riquelme, J.C. Temporal Convolutional Networks Applied to Energy-related Time Series Forecasting. Applied Sciences. 2020; 10(7):2322 (doi: 10.3390/app10072322).

We have conducted experiments for energy-related time series forecasting with diferent TCN architectures and have been compared with LSTM architectures.

Both datasets can be found at Red Electrica de España (REE).

For the TCN implementation, @philipperemy library has been used https://github.com/philipperemy/keras-tcn.