This repository contains a collection of materials needed to reproduce the results reported in the paper BaLNet: A probabilistic Bayesian neural network for solar wind speed forecasting
.
The repository has the following structure:
- Data: Contains all data files required to reproduce results. Solar wind data is provided after initial cleaning of unphysical values.
- Code: Contains the code needed to generate BaLNet. Although the
.ipynb
file us provided, this notebook was developed using Google Colab. It is therefore recommended to upload and run it in that same system to ensure compatibility. It should also be kept in mind that, for ease of access, it assumes certain information to be kept in the user's Google Drive storage. This is done in the interest of reducing data transfer times, but can be removed in favor of manual upload of the required files.