ESP  0.1
The Example-based Sensor Predictions (ESP) system tries to bring machine learning to the maker community.
user_capacitive_sensing.cpp

Capacitve sensing.

#include <ESP.h>
GestureRecognitionPipeline pipeline;
void setup() {
//pipeline.addPreProcessingModule(MovingAverageFilter(5, 3));
//pipeline.addFeatureExtractionModule(TimeDomainFeatures(10, 1, 3, false, true, true, false, false));
//pipeline.addPreProcessingModule(Derivative(Derivative::FIRST_DERIVATIVE, 0.1, 12));
pipeline.setClassifier(ANBC(false, true, 10.0)); // use scaling, use null rejection, null rejection parameter
// null rejection parameter is multiplied by the standard deviation to determine
// the rejection threshold. the higher the number, the looser the filter; the
// lower the number, the tighter the filter.
}