Branch used for personal experiments
This repository contains the code used for different pubblications. Different branch corresponds to the code used for specific articles or/and up to a specific moment in time. Below there is a list of branches with a short description for each one. The complete list of articles is at the bottom of the readme. Index of the README.
- main : default brach
- code_ICT4AWE_2023 : Code used for [1] and [2]. Contains the code for vEEGNet-ver1 and vEEGNet-ver2
- Before_rewriting : The branch contains the codes before a complete rewriting and reorganization of the repository
- backup_after_dwt_implementation : As the name suggests, it contains the codes after a backup of the repository after the implementation of DTW loss function.
- hvEEGNet_paper : Code used for [[3]][hvEEGNet_preprint]. Contains the code for vEEGNet-ver3 and hvEEGNet
- jesus-experiment : Branch for personal experiments
The code is organized as a python package inside the library folder. Inside library there are several submodules, each one dedicated to a specific purpose. Therefore the structure of the import will have the following syntax.
from library.subpackage import stuff_from_subpackageThe complete list of subpackages is :
- model : contains the definition of all the models developed. More information in the model README
- training : contains function to train the various model. More information in the training README. If you use wandb there are version of the training scripts with support for this awesome library. If you don't use wandb I highly recommend you try using it.
- config : contains the configuration for the creation of models and training. More info in the model REAMDE and the training README
- dataset: contains functions used to download dataset and to perform some basic preprocess. More info in the dataset README
If you use this repository cite [3]
Click to expand!
- [1] Zancanaro, A., Zoppis, I., Manzoni, S., & Cisotto, G. (2023). vEEGNet: A New Deep Learning Model to Classify and Generate EEG. In Proceedings of the 9th International Conference on Information and Communication Technologies for Ageing Well and e-Health, ICT4AWE 2023, Prague, Czech Republic, April 22-24, 2023 (Vol. 2023, pp. 245-252). Science and Technology Publications.
- [2] Zancanaro, A., Cisotto, G. Zoppis, I., & Manzoni, S. (2023). vEEGNet: A New Deep Learning Model to Classify and Generate EEG., vEEGNet: learning latent representations to reconstruct EEG raw data via variational autoencoders (under review) (preprint on ResearchGate)
- [3] Cisotto, Giulia and Zancanaro, Alberto and Zoppis, Italo and Manzoni, Sara, HvEEGNet: A New Deep Learning Model for High-Fidelity EEG Reconstruction. Available at SSRN: https://ssrn.com/abstract=4725025 or http:https://dx.doi.org/10.2139/ssrn.4725025 (under review) (preprint on SSRN)