signal processing

My MSc thesis work. I essayed with a number of classifiers for two type of health-related time series: electrocardiogram and blood oxygen levels. The four approaches used on each type of signal were: (i) featured based approach, where integral and differential attributes where devised and used to classify the sets (ii) Direct Machine Learning application to the time series (Perceptron, Random Forest and others) (iii) 1D convolutional neural networks applied to the signals (iv) 2D convolutional neural networks applied to the signal spectrograms.

The thesis pdf file, included here, is written in Spanish.

Python package to analyze and process time-discrete, equidistant measured signals, and visualize with Plotly

Experiments for the paper: "Improving Machine Hearing on Limited Data Sets"

Comparison of standard Gabor transform with Gabor scattering audio signal representations used as an input of convolutional neural network on multiclass classification task. Experiments conducted on synthetic and GoodSounds data.

Package for computation of Gabor scattering and Mel scattering transform.