Copiste  0.1
 All Classes Functions Variables Enumerations Friends Pages
Public Member Functions
FeatureExtractor Class Reference

An abstract class, used as an interface between the processing unit and the features algorithms. More...

#include <extractor.h>

Inheritance diagram for FeatureExtractor:
Filter HZCRRExtr LSTERExtr SpectrumExtr STEExtr ZCRExtr AverageFilter CentroidFilter CompareFilter DistanceFilter FluxFilter HighLowRatioFilter MelFilter MemoryFilter RangeFilter

List of all members.

Public Member Functions

 FeatureExtractor (int chunkSize=0)
 Constructor. Can set the typical chunk size.
virtual ~FeatureExtractor ()
 Virtual destructor : enables each extractor to delete arrays.
virtual bool extract (std::deque< uint16_t > data, int size)=0
 Run the algorithm and store the results.
bool extract (std::deque< uint16_t > data, int size, size_t start)
 Idem, but the buffer starts at index start instead (circular buffer)
virtual float value (int index=0)=0
 Retrieve the results (from the index). Some values may be out of bounds.
virtual float safeValue (int index=0)
 Retrieve the results, with the bounds respected (we apply a min / max)
virtual float value (int index, float low, float high)
 Retrieve the results, with other bounds.
virtual int size ()=0
 Get the number of available values.
virtual float min ()
 Get the minimum value outputted (we assume that each value from a single extractor is about as big as the others)
virtual float max ()
 Get the maximum value outputted.
virtual void setFloat (string key, float value)
 Set a float parameter.
virtual void setInt (string key, int value)
 Set a int parameter.
virtual float getFloat (string key)
 Get a float parameter.
virtual int getInt (string key)
 Get a int parameter.
virtual void setString (string key, string value)
 Set a string parameter.
virtual string getString (string key)
 Get a string parameter.
virtual void clear ()
 Clear the memory of the extractor (so that the previously computed results don't infer to the next ones)

Detailed Description

An abstract class, used as an interface between the processing unit and the features algorithms.


The documentation for this class was generated from the following file: