Table of Contents
AutoML BCI is a framework to train neural network dedicated for BCI Problem. it supports hyperparameters tunning using Optuna framework, and enables live tracking and model monitoring via Tensorbaord.
This framework adapts the MLOps design concept, which supports continous intergation and continous training. It is composed of three main building blocks, as illustrated in the below diagram. The building blocks of AutoML BCI are controlled by configuration files, which helps to minimizes changes in the core code, eases and simplifies the usage of the framework and most importantly allows launching multiple and parallel experiments without losing track of the parameters records.
AutoML BCI framework is build with help of the following frameworks:
Database InformationThis database includes the de-identified EEG data from 62 healthy individuals who participated in a brain-computer interface (BCI) study https://figshare.com/articles/dataset/Human_EEG_Dataset_for_Brain-Computer_Interface_and_Meditation/13123148
- how to download Dataset BBCI(351GB) mkdir -p ~/data wget https://figshare.com/ndownloader/articles/13123148/versions/1 unzip 1
how to install miniconda3
- how to install miniconda3
mkdir -p ~/miniconda3 wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3 rm -rf ~/miniconda3/miniconda.sh ~/miniconda3/bin/conda init bash ~/miniconda3/bin/conda init zsh echo "PATH=$PATH:$HOME/miniconda3/bin" >> .bashrc source .bashrc
- how to install environment
conda env create -f cluster_env.yaml
- how to install environment and Create conda environment from environment.yaml (environment_windows.yaml for Windows OS)
conda env create -f environment.yaml
Then make sure your IDE is running on this environment. In PyCharm one has to click on Python 3.7, add an interpreter and then add an existing conda environment.
- Clone the repo
git clone git@git.tu-berlin.de:roydick1.0/BCI-PJ2021SS.git
- activate conda enviroment
source .bashrc conda activate bbcpy_env - Install missing libraries
pip3 install ...
Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/AmazingFeature) - Commit your Changes (
git commit -m 'Add some AmazingFeature') - Push to the Branch (
git push origin feature/AmazingFeature) - Open a Pull Request
- Ali Alouane (ali.alouane@campus.tu-berlin.de)
- Jana-Kira SchomberPiet
- Piet Lennart Wagner
- Philip Daniel Wilson
We want to give many thanks to our project supervisor Dr. Daniel Miklody, Oleksandr Zlatov, and the Neurotechnology group at TU Berlin.