Team Project

Validating Deep Neural Networks for Online Decoding of Motor Imagery Movements from EEG Signals

Article: https://www.mdpi.com/1424-8220/19/1/210/htm
Github: https://github.com/gumpy-bci/gumpy-deeplearning/tree/master/models
Data: http://gumpy.org/
Summary: Decoding motor imagery. Used three deep learning models 1) A long short-term memory (LSTM); (2) a spectrogram-based convolutional neural network model (CNN); and (3) a recurrent convolutional neural ne$twork (RCNN). Did not use feature engineering.
Preprocessed: Gumpy gives a the data! It also gives a lot of code to preprocess
Notebook: https://colab.research.google.com/drive/12YtbtCF7jbqUk_Rtz38ZkS-94V-EtGGP
Cons: None so far!
Pros: Seems like a great paper! Very clear and has all the data, code.

Preprocessing consists of:

Name		Name	Last commit message	Last commit date
Latest commit History 50 Commits
.ipynb_checkpoints		.ipynb_checkpoints
experiments		experiments
gumpy-deeplearning		gumpy-deeplearning
gumpy-models		gumpy-models
gumpy		gumpy
Gumpy-Testing.ipynb		Gumpy-Testing.ipynb
Gumpy_Testing_EB.ipynb		Gumpy_Testing_EB.ipynb
Gumpy_Testing_SH.ipynb		Gumpy_Testing_SH.ipynb
Project_Report_SH.ipynb		Project_Report_SH.ipynb
README.md		README.md
Test_Notebook_Vilde_RCNN.ipynb		Test_Notebook_Vilde_RCNN.ipynb
Test_Notebook_Vilde_RNN_.ipynb		Test_Notebook_Vilde_RNN_.ipynb