Mercurial > hg > svcore
view transform/FeatureExtractionModelTransformer.h @ 573:bdc9bb371a9f
...
author | Chris Cannam |
---|---|
date | Fri, 13 Mar 2009 17:53:43 +0000 |
parents | 1d7ebc05157e |
children | d373ced0e6e9 5295bdb58840 |
line wrap: on
line source
/* -*- c-basic-offset: 4 indent-tabs-mode: nil -*- vi:set ts=8 sts=4 sw=4: */ /* Sonic Visualiser An audio file viewer and annotation editor. Centre for Digital Music, Queen Mary, University of London. This file copyright 2006 Chris Cannam and QMUL. This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. See the file COPYING included with this distribution for more information. */ #ifndef _FEATURE_EXTRACTION_PLUGIN_TRANSFORMER_H_ #define _FEATURE_EXTRACTION_PLUGIN_TRANSFORMER_H_ #include "ModelTransformer.h" #include <QString> #include <vamp-hostsdk/Plugin.h> #include <iostream> class DenseTimeValueModel; class FeatureExtractionModelTransformer : public ModelTransformer { Q_OBJECT public: FeatureExtractionModelTransformer(Input input, const Transform &transform); virtual ~FeatureExtractionModelTransformer(); protected: virtual void run(); Vamp::Plugin *m_plugin; Vamp::Plugin::OutputDescriptor *m_descriptor; int m_outputFeatureNo; void createOutputModel(); void addFeature(size_t blockFrame, const Vamp::Plugin::Feature &feature); void setCompletion(int); void getFrames(int channelCount, long startFrame, long size, float **buffer); // just casts DenseTimeValueModel *getConformingInput(); template <typename ModelClass> bool isOutput() { return dynamic_cast<ModelClass *>(m_output) != 0; } template <typename ModelClass> ModelClass *getConformingOutput() { ModelClass *mc = dynamic_cast<ModelClass *>(m_output); if (!mc) { std::cerr << "FeatureExtractionModelTransformer::getOutput: Output model not conformable" << std::endl; } return mc; } }; #endif