This repository contains code for energy consumption forecasting based on time series data using JITtrans model.
This project applies JITtrans architectures to forecast time series data (energy consumption).
The main code handles data preprocessing, models training, and evaluation, while the model.py file contains the model architecture.
The source code is related to the paper "Just In Time Transformers", by AHMED ALA EDDINE BENALI, MASSIMO CAFARO, ITALO EPICOCO, MARCO PULIMENO and ENRICO JUNIOR SCHIOPPA (submitted)
main.py: Main code for loading data, training, and evaluating the model.
model.py: Model code defining the Transformer model.
requirements.txt: List of dependencies for the project.
Make sure you have Python installed. You can download it here.
To install all the required dependencies, use the following command: pip install -r requirements.txt