Pinned Repositories
astroML
Machine learning, statistics, and data mining for astronomy and astrophysics
Automatic-Seizure-Detection-from-EEG-Hilbert-Marginal-Spectrum---Reproducible-Research
ECE-603 Statistical Signal Processing - Course Project
catboost
A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.
CodePaper_JCAP_2020
ABSTRACT: This code is an example of a statistical analysis that aims to obtain new and tight estimates on Hubble parameter, obtained by considering two data sets from galaxy distribution observations: galaxy cluster gas mass fractions and baryon acoustic oscillation measurements.
demo-repo
demo for github intro video
EEG-motor-imagery
ECE-GY 9123 Project: GCN-Explain-Net: An Explainable Graph Convolutional Neural Network (GCN) for EEG-based Motor Imagery Classification and Demystification
EEG_Signals_GAMEEMO
EEG_to_-CWT
Convert EEG signals to images using Continuous Wavelet Transform.
EMCMC
Evaluation-of-Boosting-Algorithms-for-P300-Detection-in-EEG-Signals
Brain-Computer Interfaces have impacted the lives of many, especially those whose mobility and ability to speak are affected. It is able to do so by bridging the gap between thoughts and devices. One of its most popular applications, the P300 Speller, is a powerful aid that allows the patients to regain a certain level of autonomy. Detection of a P300 peak and character identification are the two major components of a P300 speller. In this study, the first component of P300 speller is covered. Various conventional learning algorithms like Support Vector Machine, Discriminant Analysis, Neural Network and their variants have been used in previous studies. These methods have limitations: some are prone to overfitting; others require a large amount of training data, while there are some limitations that necessitate complicated computing thus making them less favorable for real-time analysis. Boosting algorithms are very less explored in the field of Electroencephalography (EEG) and less prone to most of the limitations of these conventional models. This paper evaluates the performances of LightGBM and CatBoost on the dataset used in the competition BCI NER 2015 on Kaggle. These algorithms have recently gained popularity and have proven to be powerful. Further, they are compared with the performances of XGBoost and AdaBoost and a maximum 1 Score of 0.84 was achieved using LightGBM as a classifier.
Bethany-Gosala's Repositories
Bethany-Gosala/EEG_Signals_GAMEEMO
Bethany-Gosala/astroML
Machine learning, statistics, and data mining for astronomy and astrophysics
Bethany-Gosala/Automatic-Seizure-Detection-from-EEG-Hilbert-Marginal-Spectrum---Reproducible-Research
ECE-603 Statistical Signal Processing - Course Project
Bethany-Gosala/catboost
A fast, scalable, high performance Gradient Boosting on Decision Trees library, used for ranking, classification, regression and other machine learning tasks for Python, R, Java, C++. Supports computation on CPU and GPU.
Bethany-Gosala/CodePaper_JCAP_2020
ABSTRACT: This code is an example of a statistical analysis that aims to obtain new and tight estimates on Hubble parameter, obtained by considering two data sets from galaxy distribution observations: galaxy cluster gas mass fractions and baryon acoustic oscillation measurements.
Bethany-Gosala/demo-repo
demo for github intro video
Bethany-Gosala/EEG-motor-imagery
ECE-GY 9123 Project: GCN-Explain-Net: An Explainable Graph Convolutional Neural Network (GCN) for EEG-based Motor Imagery Classification and Demystification
Bethany-Gosala/EEG_to_-CWT
Convert EEG signals to images using Continuous Wavelet Transform.
Bethany-Gosala/EMCMC
Bethany-Gosala/Evaluation-of-Boosting-Algorithms-for-P300-Detection-in-EEG-Signals
Brain-Computer Interfaces have impacted the lives of many, especially those whose mobility and ability to speak are affected. It is able to do so by bridging the gap between thoughts and devices. One of its most popular applications, the P300 Speller, is a powerful aid that allows the patients to regain a certain level of autonomy. Detection of a P300 peak and character identification are the two major components of a P300 speller. In this study, the first component of P300 speller is covered. Various conventional learning algorithms like Support Vector Machine, Discriminant Analysis, Neural Network and their variants have been used in previous studies. These methods have limitations: some are prone to overfitting; others require a large amount of training data, while there are some limitations that necessitate complicated computing thus making them less favorable for real-time analysis. Boosting algorithms are very less explored in the field of Electroencephalography (EEG) and less prone to most of the limitations of these conventional models. This paper evaluates the performances of LightGBM and CatBoost on the dataset used in the competition BCI NER 2015 on Kaggle. These algorithms have recently gained popularity and have proven to be powerful. Further, they are compared with the performances of XGBoost and AdaBoost and a maximum 1 Score of 0.84 was achieved using LightGBM as a classifier.
Bethany-Gosala/Extracting-Image-from-EEG-signals
Bethany-Gosala/fusion-gcn
Bethany-Gosala/gpubootcamp
This repository consists for gpu bootcamp material for HPC and AI
Bethany-Gosala/Hubble
Estimating the Age of universe using galaxies distance and velocity data
Bethany-Gosala/Hubble_Constant
Calculating hubble constant
Bethany-Gosala/kymatio
Wavelet scattering transforms in Python with GPU acceleration
Bethany-Gosala/MedFuse
Bethany-Gosala/mne-features
MNE-Features software for extracting features from multivariate time series
Bethany-Gosala/models_K2
Models built with TensorFlow
Bethany-Gosala/observational-cosmology
Exercises of astrophysics IV course at EPFL
Bethany-Gosala/RosslerAttractorDynamicalSystems
Plot a 2D and 3D attractor using Runge-Kutta method.
Bethany-Gosala/Talhaanwarch-youtube-tutorials
Bethany-Gosala/TensorFlow-Pokemon-Course