Skip to content
No description, website, or topics provided.
Jupyter Notebook HTML Other
Branch: master
Clone or download

Latest commit

Fetching latest commit…
Cannot retrieve the latest commit at this time.

Files

Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Report init rapport Dec 17, 2015
data helper pour charger les données, mise à jour de dataset exploration e… Oct 19, 2015
features J'ai rajouté la librairie python-speech-features comme ça vous avez pas Nov 18, 2015
presentation repartition des slides Oct 29, 2015
results-cnn best model Jan 12, 2016
tools ajout deterministic dataset shuffle pour uniformiser la répartition d… Oct 26, 2015
.gitignore ajout pyc dans gitignore Oct 19, 2015
Classifieur - Amin - SVN.ipynb SVN binaire et multiclass Nov 17, 2015
Classifieur - Marc - Butterworth + k-nn.ipynb Best Result : 94,08% Dec 21, 2015
Classifieur-Vincent-MFCC+NaifBayes.ipynb Le premier jet du Ultimate Classifier Nov 26, 2015
Classifieurs - Amin - arbre de décision .ipynb classifieur L'arbre de décision Dec 21, 2015
Classifieurs - Thomas - CNN (binaire, 98%).ipynb ajout notebook cnn 98% Nov 10, 2015
Classifieurs - Thomas - Tests Fourier + PCA pour classification binaire i vs j pour tout i et j.ipynb ajout i vs j pour tout i et j Nov 27, 2015
Classifieurs - Thomas - Tests Fourier + PCA.ipynb ajout tests fourier + pca (classification binaire avec 90% de bons ré… Nov 6, 2015
Classifieurs - Thomas - Tests std dev + sklearn.ipynb fausse alerte il y avait une typo (precision = 9.5%) Oct 29, 2015
Classifieurs_René_FFT_PCA_GPC_KNN.ipynb FFT-PCA-(GPC,75%)-K-PPv,98% Nov 16, 2015
Dataset exploration - Grasp and lift.ipynb ajout early version du dataset explo pour grasp and lift Dec 19, 2015
Dataset exploration.ipynb ajout deterministic dataset shuffle pour uniformiser la répartition d… Oct 26, 2015
Présentation_nov_2015.pdf ajout présentation Nov 12, 2015
README.md readme Nov 12, 2015
UltimateFeatureVectorClassifierFromHell.ipynb Le premier jet du Ultimate Classifier Nov 26, 2015

README.md

Giving sense to EEG records

As a project during the course IFT6390 "machine learning" by Pascal Vincent from MILA, we have been working on the dataset Mind big data provided by David Vivancos. This repository groups several tests and documents we produced during our work.

Mind big data

[http://www.mindbigdata.com/]

We have focused our work on Mind big data. This dataset, provided by David Vivancos, is composed of records of 2 seconds of EEG signals using several non-medical grade headsets. During the experiment, the subject was presented digits (0 - 9) on a computer, and tried to focus on the digit. We aim at discovering if we can predict which digit was presented using the sensors series.

Usage

If you wish to reproduce our experiments, you can load the ipython notebooks and run the code.

You will require several libraries: numpy and sklearn for most notebooks and Theano for CNN related notebooks.

Team members

  • Amin Shahab
  • Marc Sayn-Urpar
  • René Doumbouya
  • Thomas George
  • Vincent Antaki
You can’t perform that action at this time.