annotate vamp-sdk/hostext/PluginSummarisingAdapter.cpp @ 185:701505ac170c

* Reorganise accumulators to facilitate segmentation

/* -*- c-basic-offset: 4 indent-tabs-mode: nil -*-  vi:set ts=8 sts=4 sw=4: */

/*
    Vamp

    An API for audio analysis and feature extraction plugins.

    Centre for Digital Music, Queen Mary, University of London.
    Copyright 2006-2008 Chris Cannam and QMUL.
  
    Permission is hereby granted, free of charge, to any person
    obtaining a copy of this software and associated documentation
    files (the "Software"), to deal in the Software without
    restriction, including without limitation the rights to use, copy,
    modify, merge, publish, distribute, sublicense, and/or sell copies
    of the Software, and to permit persons to whom the Software is
    furnished to do so, subject to the following conditions:

    The above copyright notice and this permission notice shall be
    included in all copies or substantial portions of the Software.

    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
    EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
    MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
    NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
    ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
    CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
    WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

    Except as contained in this notice, the names of the Centre for
    Digital Music; Queen Mary, University of London; and Chris Cannam
    shall not be used in advertising or otherwise to promote the sale,
    use or other dealings in this Software without prior written
    authorization.
*/

#include "PluginSummarisingAdapter.h"

#include <map>
#include <cmath>
#include <climits>

namespace Vamp {

namespace HostExt {

class PluginSummarisingAdapter::Impl
{
public:
    Impl(Plugin *plugin, float inputSampleRate);
    ~Impl();

    FeatureSet process(const float *const *inputBuffers, RealTime timestamp);
    FeatureSet getRemainingFeatures();

    void setSummarySegmentBoundaries(const SegmentBoundaries &);

    FeatureList getSummaryForOutput(int output,
                                    SummaryType type,
                                    AveragingMethod avg);

    FeatureSet getSummaryForAllOutputs(SummaryType type,
                                       AveragingMethod avg);

protected:
    Plugin *m_plugin;
    float m_inputSampleRate;

    SegmentBoundaries m_boundaries;

    typedef std::vector<float> ValueList;

    struct Result { // smaller than Feature
        RealTime time;
        RealTime duration;
        ValueList values; // bin number -> value
    };

    typedef std::vector<Result> ResultList;

    struct OutputAccumulator {
        int bins;
        ResultList results;
        OutputAccumulator() : bins(0) { }
    };

    typedef std::map<int, OutputAccumulator> OutputAccumulatorMap;
    OutputAccumulatorMap m_accumulators; // output number -> accumulator

    typedef std::map<RealTime, OutputAccumulator> SegmentAccumulatorMap;
    typedef std::map<int, SegmentAccumulatorMap> OutputSegmentAccumulatorMap;
    OutputSegmentAccumulatorMap m_segmentedAccumulators;

    typedef std::map<int, RealTime> OutputTimestampMap;
    OutputTimestampMap m_prevTimestamps; // output number -> timestamp
    OutputTimestampMap m_prevDurations; // output number -> durations

    struct OutputBinSummary {

        int count;

        // extents
        float minimum;
        float maximum;
        float sum;

        // sample-average results
        float median;
        float mode;
        float variance;

        // continuous-time average results
        float median_c;
        float mode_c;
        float mean_c;
        float variance_c;
    };

    typedef std::map<int, OutputBinSummary> OutputSummary;
    typedef std::map<RealTime, OutputSummary> SummarySegmentMap;
    typedef std::map<int, SummarySegmentMap> OutputSummarySegmentMap;

    OutputSummarySegmentMap m_summaries;

    bool m_reduced;
    RealTime m_lastTimestamp;

    void accumulate(const FeatureSet &fs, RealTime, bool final);
    void accumulate(int output, const Feature &f, RealTime, bool final);
    void accumulateFinalDurations();
    void segment();
    void reduce();
};

static RealTime INVALID_DURATION(INT_MIN, INT_MIN);
    
PluginSummarisingAdapter::PluginSummarisingAdapter(Plugin *plugin) :
    PluginWrapper(plugin)
{
    m_impl = new Impl(plugin, m_inputSampleRate);
}

PluginSummarisingAdapter::~PluginSummarisingAdapter()
{
    delete m_impl;
}

Plugin::FeatureSet
PluginSummarisingAdapter::process(const float *const *inputBuffers, RealTime timestamp)
{
    return m_impl->process(inputBuffers, timestamp);
}

Plugin::FeatureSet
PluginSummarisingAdapter::getRemainingFeatures()
{
    return m_impl->getRemainingFeatures();
}

Plugin::FeatureList
PluginSummarisingAdapter::getSummaryForOutput(int output,
                                              SummaryType type,
                                              AveragingMethod avg)
{
    return m_impl->getSummaryForOutput(output, type, avg);
}

Plugin::FeatureSet
PluginSummarisingAdapter::getSummaryForAllOutputs(SummaryType type,
                                                  AveragingMethod avg)
{
    return m_impl->getSummaryForAllOutputs(type, avg);
}

PluginSummarisingAdapter::Impl::Impl(Plugin *plugin, float inputSampleRate) :
    m_plugin(plugin),
    m_inputSampleRate(inputSampleRate),
    m_reduced(false)
{
}

PluginSummarisingAdapter::Impl::~Impl()
{
}

Plugin::FeatureSet
PluginSummarisingAdapter::Impl::process(const float *const *inputBuffers, RealTime timestamp)
{
    if (m_reduced) {
        std::cerr << "WARNING: Cannot call PluginSummarisingAdapter::process() or getRemainingFeatures() after one of the getSummary methods" << std::endl;
    }
    FeatureSet fs = m_plugin->process(inputBuffers, timestamp);
    accumulate(fs, timestamp, false);
    //!!! should really be "timestamp plus step size"
    m_lastTimestamp = timestamp;
    return fs;
}

Plugin::FeatureSet
PluginSummarisingAdapter::Impl::getRemainingFeatures()
{
    if (m_reduced) {
        std::cerr << "WARNING: Cannot call PluginSummarisingAdapter::process() or getRemainingFeatures() after one of the getSummary methods" << std::endl;
    }
    FeatureSet fs = m_plugin->getRemainingFeatures();
    accumulate(fs, m_lastTimestamp, true);
    return fs;
}

Plugin::FeatureList
PluginSummarisingAdapter::Impl::getSummaryForOutput(int output,
                                                    SummaryType type,
                                                    AveragingMethod avg)
{
    if (!m_reduced) {
        segment();
        reduce();
        m_reduced = true;
    }

    bool continuous = (avg == ContinuousTimeAverage);

    FeatureList fl;
    for (SummarySegmentMap::const_iterator i = m_summaries[output].begin();
         i != m_summaries[output].end(); ++i) {

        Feature f;
        f.hasTimestamp = true;
        f.timestamp = i->first;
        f.hasDuration = false;

        for (OutputSummary::const_iterator j = i->second.begin();
             j != i->second.end(); ++j) {

            // these will be ordered by bin number, and no bin numbers
            // will be missing except at the end (because of the way
            // the accumulators were initially filled in accumulate())

            const OutputBinSummary &summary = j->second;
            float result = 0.f;

            switch (type) {

            case Minimum:
                result = summary.minimum;
                break;

            case Maximum:
                result = summary.maximum;
                break;

            case Mean:
                if (continuous) {
                    result = summary.mean_c;
                } else if (summary.count) {
                    result = summary.sum / summary.count;
                }
                break;

            case Median:
                if (continuous) result = summary.median_c;
                else result = summary.median;
                break;

            case Mode:
                if (continuous) result = summary.mode_c;
                else result = summary.mode;
                break;

            case Sum:
                result = summary.sum;
                break;

            case Variance:
                if (continuous) result = summary.variance_c;
                else result = summary.variance;
                break;

            case StandardDeviation:
                if (continuous) result = sqrtf(summary.variance_c);
                else result = sqrtf(summary.variance);
                break;

            case Count:
                result = summary.count;
                break;

            default:
                break;
            }
            
            f.values.push_back(result);
        }

        fl.push_back(f);
    }
    return fl;
}

Plugin::FeatureSet
PluginSummarisingAdapter::Impl::getSummaryForAllOutputs(SummaryType type,
                                                        AveragingMethod avg)
{
    if (!m_reduced) {
        segment();
        reduce();
        m_reduced = true;
    }

    FeatureSet fs;
    for (OutputSummarySegmentMap::const_iterator i = m_summaries.begin();
         i != m_summaries.end(); ++i) {
        fs[i->first] = getSummaryForOutput(i->first, type, avg);
    }
    return fs;
}

void
PluginSummarisingAdapter::Impl::accumulate(const FeatureSet &fs,
                                           RealTime timestamp, 
                                           bool final)
{
    for (FeatureSet::const_iterator i = fs.begin(); i != fs.end(); ++i) {
        for (FeatureList::const_iterator j = i->second.begin();
             j != i->second.end(); ++j) {
            if (j->hasTimestamp) {
                accumulate(i->first, *j, j->timestamp, final);
            } else {
                //!!! is this correct?
                accumulate(i->first, *j, timestamp, final);
            }
        }
    }
}

void
PluginSummarisingAdapter::Impl::accumulate(int output,
                                           const Feature &f,
                                           RealTime timestamp,
                                           bool final)
{
//!!! to do: use timestamp to determine which segment we're on
    
//!!! What should happen if a feature's duration spans a segment
// boundary?  I think we probably want to chop it, and pretend that it
// appears in both -- don't we? do we?  A very long feature (e.g. key,
// if the whole audio is in a single key) might span many or all
// segments, and we want that to be reflected in the results (e.g. it
// is the modal key in all of those segments, not just the first).
// That is actually quite complicated to do!

//!!! This affects how we record things.  If features spanning a
// boundary should be chopped, then we need to have per-segment
// accumulators (and the feature value goes into both -- perhaps we
// need a separate phase to split the accumulator up into segments).
// If features spanning a boundary should be counted only in the first
// segment, with their full duration, then we should store them in a
// single accumulator and distribute into segments only on reduce.

    std::cerr << "output " << output << ": timestamp " << timestamp << ", prev timestamp " << m_prevTimestamps[output] << ", final " << final << std::endl;

    // At each process step, accumulate() is called once for each
    // feature on each output within that process's returned feature
    // list, and with the timestamp passed in being that of the start
    // of the process block.

    // At the end (in getRemainingFeatures), accumulate() is called
    // once for each feature on each output within the feature list
    // returned by getRemainingFeatures, and with the timestamp being
    // the same as the last process block and final set to true.

    // (What if getRemainingFeatures doesn't return any features?  We
    // still need to ensure that the final duration is written.  Need
    // a separate function to close the durations.)

    // At each call, we pull out the value for the feature and stuff
    // it into the accumulator's appropriate values array; and we
    // calculate the duration for the _previous_ feature, or pull it
    // from the prevDurations array if the previous feature had a
    // duration in its structure, and stuff that into the
    // accumulator's appropriate durations array.

    if (m_prevDurations.find(output) != m_prevDurations.end()) {

        // Not the first time accumulate has been called for this
        // output -- there has been a previous feature

        RealTime prevDuration;

        // Note that m_prevDurations[output] only contains the
        // duration field that was contained in the previous feature.
        // If it didn't have an explicit duration,
        // m_prevDurations[output] should be INVALID_DURATION and we
        // will have to calculate the duration from the previous and
        // current timestamps.

        if (m_prevDurations[output] != INVALID_DURATION) {
            prevDuration = m_prevDurations[output];
            std::cerr << "Previous duration from previous feature: " << prevDuration << std::endl;
        } else {
            prevDuration = timestamp - m_prevTimestamps[output];
            std::cerr << "Previous duration from diff: " << timestamp << " - "
                      << m_prevTimestamps[output] << std::endl;
        }

        std::cerr << "output " << output << ": ";

        std::cerr << "Pushing previous duration as " << prevDuration << std::endl;
        
        m_accumulators[output].results
            [m_accumulators[output].results.size() - 1]
            .duration = prevDuration;
    }

    if (f.hasDuration) m_prevDurations[output] = f.duration;
    else m_prevDurations[output] = INVALID_DURATION;

    m_prevTimestamps[output] = timestamp;

    //!!! should really be "timestamp plus duration" or "timestamp plus output resolution"
    if (timestamp > m_lastTimestamp) m_lastTimestamp = timestamp;

    Result result;
    result.time = timestamp;
    result.duration = INVALID_DURATION;

    if (f.values.size() > m_accumulators[output].bins) {
        m_accumulators[output].bins = f.values.size();
    }

    for (int i = 0; i < int(f.values.size()); ++i) {
        result.values.push_back(f.values[i]);
    }

    m_accumulators[output].results.push_back(result);
}

void
PluginSummarisingAdapter::Impl::accumulateFinalDurations()
{
    for (OutputTimestampMap::iterator i = m_prevTimestamps.begin();
         i != m_prevTimestamps.end(); ++i) {

        int output = i->first;

        int acount = m_accumulators[output].results.size();

        if (acount == 0) continue;

        RealTime prevTimestamp = i->second;

        std::cerr << "output " << output << ": ";

        if (m_prevDurations.find(output) != m_prevDurations.end() &&
            m_prevDurations[output] != INVALID_DURATION) {

            std::cerr << "Pushing final duration from feature as " << m_prevDurations[output] << std::endl;

            m_accumulators[output].results[acount - 1].duration =
                m_prevDurations[output];

        } else {

            std::cerr << "Pushing final duration from diff as " << m_lastTimestamp << " - " << m_prevTimestamps[output] << std::endl;

            m_accumulators[output].results[acount - 1].duration =
                m_lastTimestamp - m_prevTimestamps[output];
        }
    }
}

void
PluginSummarisingAdapter::Impl::segment()
{
/*
    SegmentBoundaries::iterator boundaryitr = m_boundaries.begin();
    RealTime segmentStart = RealTime::zeroTime;

    for (OutputAccumulatorMap::iterator i = m_accumulators.begin();
         i != m_accumulators.end(); ++i) {

        int output = i->first;
        OutputAccumulator &source = i->second;
        RealTime accumulatedTime = RealTime::zeroTime;

        for (int n = 0; n < source.durations.size(); ++n) {
*/          
            

/*
        if (boundaryitr == m_boundaries.end()) {
            m_segmentedAccumulators[output][segmentStart] = source;
            source.clear();
            continue;
        }
*/
        
            

        
}

struct ValueDurationFloatPair
{
    float value;
    float duration;

    ValueDurationFloatPair() : value(0), duration(0) { }
    ValueDurationFloatPair(float v, float d) : value(v), duration(d) { }
    ValueDurationFloatPair &operator=(const ValueDurationFloatPair &p) {
        value = p.value;
        duration = p.duration;
        return *this;
    }
    bool operator<(const ValueDurationFloatPair &p) const {
        return value < p.value;
    }
};

static double toSec(const RealTime &r)
{
    return r.sec + double(r.nsec) / 1000000000.0;
}

void
PluginSummarisingAdapter::Impl::reduce()
{
    accumulateFinalDurations();

    RealTime segmentStart = RealTime::zeroTime; //!!!

    for (OutputAccumulatorMap::iterator i = m_accumulators.begin();
         i != m_accumulators.end(); ++i) {

        int output = i->first;
        OutputAccumulator &accumulator = i->second;

        int sz = accumulator.results.size();

        double totalDuration = 0.0;
        //!!! is this right?
        if (sz > 0) {
            totalDuration = toSec(accumulator.results[sz-1].time +
                                  accumulator.results[sz-1].duration);
        }

        for (int bin = 0; bin < accumulator.bins; ++bin) {

            // work on all values over time for a single bin

            OutputBinSummary summary;

            summary.count = sz;

            summary.minimum = 0.f;
            summary.maximum = 0.f;

            summary.median = 0.f;
            summary.mode = 0.f;
            summary.sum = 0.f;
            summary.variance = 0.f;

            summary.median_c = 0.f;
            summary.mode_c = 0.f;
            summary.mean_c = 0.f;
            summary.variance_c = 0.f;

            if (sz == 0) continue;

            std::vector<ValueDurationFloatPair> valvec;

            for (int k = 0; k < sz; ++k) {
                while (accumulator.results[k].values.size() <
                       accumulator.bins) {
                    accumulator.results[k].values.push_back(0.f);
                }
            }

            for (int k = 0; k < sz; ++k) {
                float value = accumulator.results[k].values[bin];
                valvec.push_back(ValueDurationFloatPair
                                 (value,
                                  toSec(accumulator.results[k].duration)));
            }

            std::sort(valvec.begin(), valvec.end());

            summary.minimum = valvec[0].value;
            summary.maximum = valvec[sz-1].value;

            if (sz % 2 == 1) {
                summary.median = valvec[sz/2].value;
            } else {
                summary.median = (valvec[sz/2].value + valvec[sz/2 + 1].value) / 2;
            }
            
            double duracc = 0.0;
            summary.median_c = valvec[sz-1].value;

            for (int k = 0; k < sz; ++k) {
                duracc += valvec[k].duration;
                if (duracc > totalDuration/2) {
                    summary.median_c = valvec[k].value;
                    break;
                }
            }

            std::cerr << "median_c = " << summary.median_c << std::endl;
            std::cerr << "median = " << summary.median << std::endl;
                
            std::map<float, int> distribution;

            for (int k = 0; k < sz; ++k) {
                summary.sum += accumulator.results[k].values[bin];
                distribution[accumulator.results[k].values[bin]] += 1;
            }

            int md = 0;

            for (std::map<float, int>::iterator di = distribution.begin();
                 di != distribution.end(); ++di) {
                if (di->second > md) {
                    md = di->second;
                    summary.mode = di->first;
                }
            }

            distribution.clear();

            std::map<float, double> distribution_c;

            for (int k = 0; k < sz; ++k) {
                distribution_c[accumulator.results[k].values[bin]]
                    += toSec(accumulator.results[k].duration);
            }

            double mrd = 0.0;

            for (std::map<float, double>::iterator di = distribution_c.begin();
                 di != distribution_c.end(); ++di) {
                if (di->second > mrd) {
                    mrd = di->second;
                    summary.mode_c = di->first;
                }
            }

            distribution_c.clear();

            if (totalDuration > 0.0) {

                double sum_c = 0.0;

                for (int k = 0; k < sz; ++k) {
                    double value = accumulator.results[k].values[bin]
                        * toSec(accumulator.results[k].duration);
                    sum_c += value;
                }

                std::cerr << "mean_c = " << sum_c << " / " << totalDuration << " = "
                          << sum_c / totalDuration << " (sz = " << sz << ")" << std::endl;
                
                summary.mean_c = sum_c / totalDuration;

                for (int k = 0; k < sz; ++k) {
                    double value = accumulator.results[k].values[bin]
                        * toSec(accumulator.results[k].duration);
                    summary.variance_c +=
                        (value - summary.mean_c) * (value - summary.mean_c);
                }

                summary.variance_c /= summary.count;
            }

            float mean = summary.sum / summary.count;

            std::cerr << "mean = " << summary.sum << " / " << summary.count << " = "
                      << summary.sum / summary.count << std::endl;

            for (int k = 0; k < sz; ++k) {
                float value = accumulator.results[k].values[bin];
                summary.variance += (value - mean) * (value - mean);
            }
            summary.variance /= summary.count;

            m_summaries[output][segmentStart][bin] = summary;
        }
    }

    m_accumulators.clear();
}


}

}