diff --git a/bob/learn/misc/cpp/KMeansTrainer.cpp b/bob/learn/misc/cpp/KMeansTrainer.cpp
index 092c87d48698d46e4a45d1ac03a0c4337b5ed04a..583543f7bbc1b249a435cc3242fabbf14c251e84 100644
--- a/bob/learn/misc/cpp/KMeansTrainer.cpp
+++ b/bob/learn/misc/cpp/KMeansTrainer.cpp
@@ -12,43 +12,73 @@
#include <boost/random.hpp>
+/*
bob::learn::misc::KMeansTrainer::KMeansTrainer(double convergence_threshold,
- size_t max_iterations, bool compute_likelihood, InitializationMethod i_m):
- bob::learn::misc::EMTrainer<bob::learn::misc::KMeansMachine, blitz::Array<double,2> >(
- convergence_threshold, max_iterations, compute_likelihood),
- m_initialization_method(i_m),
- m_rng(new boost::mt19937()), m_average_min_distance(0),
- m_zeroethOrderStats(0), m_firstOrderStats(0,0)
+ size_t max_iterations, bool compute_likelihood, InitializationMethod i_m)
{
+
+ m_initialization_method = i_m;
+ m_zeroethOrderStats = 0;
+ m_firstOrderStats = 0;
+ m_average_min_distance = 0;
+
+ m_compute_likelihood = compute_likelihood;
+ m_convergence_threshold = convergence_threshold;
+ m_max_iterations = max_iterations;
+ //m_rng(new boost::mt19937());
+
}
+*/
-bob::learn::misc::KMeansTrainer::KMeansTrainer(const bob::learn::misc::KMeansTrainer& other):
- bob::learn::misc::EMTrainer<bob::learn::misc::KMeansMachine, blitz::Array<double,2> >(
- other.m_convergence_threshold, other.m_max_iterations, other.m_compute_likelihood),
- m_initialization_method(other.m_initialization_method),
- m_rng(other.m_rng), m_average_min_distance(other.m_average_min_distance),
- m_zeroethOrderStats(bob::core::array::ccopy(other.m_zeroethOrderStats)),
- m_firstOrderStats(bob::core::array::ccopy(other.m_firstOrderStats))
+bob::learn::misc::KMeansTrainer::KMeansTrainer(InitializationMethod i_m)
{
+
+ m_initialization_method = i_m;
+ m_zeroethOrderStats = 0;
+ m_firstOrderStats = 0;
+ m_average_min_distance = 0;
+
+ //m_rng(new boost::mt19937());
+}
+
+
+bob::learn::misc::KMeansTrainer::KMeansTrainer(const bob::learn::misc::KMeansTrainer& other){
+
+ //m_convergence_threshold = other.m_convergence_threshold;
+ //m_max_iterations = other.m_max_iterations;
+ //m_compute_likelihood = other.m_compute_likelihood;
+
+ m_initialization_method = other.m_initialization_method;
+ m_rng = other.m_rng;
+ m_average_min_distance = other.m_average_min_distance;
+ m_zeroethOrderStats = bob::core::array::ccopy(other.m_zeroethOrderStats);
+ m_firstOrderStats = bob::core::array::ccopy(other.m_firstOrderStats);
}
+
bob::learn::misc::KMeansTrainer& bob::learn::misc::KMeansTrainer::operator=
(const bob::learn::misc::KMeansTrainer& other)
{
if(this != &other)
{
- EMTrainer<bob::learn::misc::KMeansMachine, blitz::Array<double,2> >::operator=(other);
- m_initialization_method = other.m_initialization_method;
- m_rng = other.m_rng;
- m_average_min_distance = other.m_average_min_distance;
- m_zeroethOrderStats.reference(bob::core::array::ccopy(other.m_zeroethOrderStats));
- m_firstOrderStats.reference(bob::core::array::ccopy(other.m_firstOrderStats));
+ //m_compute_likelihood = other.m_compute_likelihood;
+ //m_convergence_threshold = other.m_convergence_threshold;
+ //m_max_iterations = other.m_max_iterations;
+ m_rng = other.m_rng;
+ m_initialization_method = other.m_initialization_method;
+ m_average_min_distance = other.m_average_min_distance;
+
+ m_zeroethOrderStats = bob::core::array::ccopy(other.m_zeroethOrderStats);
+ m_firstOrderStats = bob::core::array::ccopy(other.m_firstOrderStats);
}
return *this;
}
+
bool bob::learn::misc::KMeansTrainer::operator==(const bob::learn::misc::KMeansTrainer& b) const {
- return EMTrainer<bob::learn::misc::KMeansMachine, blitz::Array<double,2> >::operator==(b) &&
+ return //m_compute_likelihood == b.m_compute_likelihood &&
+ //m_convergence_threshold == b.m_convergence_threshold &&
+ //m_max_iterations == b.m_max_iterations &&
m_initialization_method == b.m_initialization_method &&
*m_rng == *(b.m_rng) && m_average_min_distance == b.m_average_min_distance &&
bob::core::array::hasSameShape(m_zeroethOrderStats, b.m_zeroethOrderStats) &&
@@ -97,7 +127,7 @@ void bob::learn::misc::KMeansTrainer::initialize(bob::learn::misc::KMeansMachine
bool valid = true;
for(size_t j=0; j<i && valid; ++j)
{
- kmeans.getMean(j, cur_mean);
+ cur_mean = kmeans.getMean(j);
valid = blitz::any(mean != cur_mean);
}
// if different, stop otherwise, try with another one
@@ -204,10 +234,6 @@ double bob::learn::misc::KMeansTrainer::computeLikelihood(bob::learn::misc::KMea
return m_average_min_distance;
}
-void bob::learn::misc::KMeansTrainer::finalize(bob::learn::misc::KMeansMachine& kmeans,
- const blitz::Array<double,2>& ar)
-{
-}
bool bob::learn::misc::KMeansTrainer::resetAccumulators(bob::learn::misc::KMeansMachine& kmeans)
{
diff --git a/bob/learn/misc/include/bob.learn.misc/KMeansTrainer.h b/bob/learn/misc/include/bob.learn.misc/KMeansTrainer.h
index 110b82ae16a5304890f5a86958978998b7567f43..caba9f64a1c2df62f7f71d344cdaa280d9af152b 100644
--- a/bob/learn/misc/include/bob.learn.misc/KMeansTrainer.h
+++ b/bob/learn/misc/include/bob.learn.misc/KMeansTrainer.h
@@ -9,8 +9,8 @@
#define BOB_LEARN_MISC_KMEANSTRAINER_H
#include <bob.learn.misc/KMeansMachine.h>
-#include <bob.learn.misc/EMTrainer.h>
#include <boost/version.hpp>
+#include <boost/random/mersenne_twister.hpp>
namespace bob { namespace learn { namespace misc {
@@ -20,7 +20,7 @@ namespace bob { namespace learn { namespace misc {
* @details See Section 9.1 of Bishop, "Pattern recognition and machine learning", 2006
* It uses a random initialisation of the means followed by the expectation-maximization algorithm
*/
-class KMeansTrainer: public EMTrainer<bob::learn::misc::KMeansMachine, blitz::Array<double,2> >
+class KMeansTrainer
{
public:
/**
@@ -40,9 +40,13 @@ class KMeansTrainer: public EMTrainer<bob::learn::misc::KMeansMachine, blitz::Ar
/**
* @brief Constructor
*/
+ KMeansTrainer(InitializationMethod=RANDOM);
+
+ /*
KMeansTrainer(double convergence_threshold=0.001,
size_t max_iterations=10, bool compute_likelihood=true,
- InitializationMethod=RANDOM);
+ InitializationMethod=RANDOM);*/
+
/**
* @brief Virtualize destructor
@@ -85,7 +89,7 @@ class KMeansTrainer: public EMTrainer<bob::learn::misc::KMeansMachine, blitz::Ar
* Data is split into as many chunks as there are means,
* then each mean is set to a random example within each chunk.
*/
- virtual void initialize(bob::learn::misc::KMeansMachine& kMeansMachine,
+ void initialize(bob::learn::misc::KMeansMachine& kMeansMachine,
const blitz::Array<double,2>& sampler);
/**
@@ -94,25 +98,21 @@ class KMeansTrainer: public EMTrainer<bob::learn::misc::KMeansMachine, blitz::Ar
* - average (Square Euclidean) distance from the closest mean
* Implements EMTrainer::eStep(double &)
*/
- virtual void eStep(bob::learn::misc::KMeansMachine& kmeans,
+ void eStep(bob::learn::misc::KMeansMachine& kmeans,
const blitz::Array<double,2>& data);
/**
* @brief Updates the mean based on the statistics from the E-step.
*/
- virtual void mStep(bob::learn::misc::KMeansMachine& kmeans,
+ void mStep(bob::learn::misc::KMeansMachine& kmeans,
const blitz::Array<double,2>&);
/**
* @brief This functions returns the average min (Square Euclidean)
* distance (average distance to the closest mean)
*/
- virtual double computeLikelihood(bob::learn::misc::KMeansMachine& kmeans);
+ double computeLikelihood(bob::learn::misc::KMeansMachine& kmeans);
- /**
- * @brief Function called at the end of the training
- */
- virtual void finalize(bob::learn::misc::KMeansMachine& kMeansMachine, const blitz::Array<double,2>& sampler);
/**
* @brief Reset the statistics accumulators
@@ -156,7 +156,13 @@ class KMeansTrainer: public EMTrainer<bob::learn::misc::KMeansMachine, blitz::Ar
void setAverageMinDistance(const double value) { m_average_min_distance = value; }
- protected:
+ private:
+
+ //bool m_compute_likelihood; ///< whether lilelihood is computed during the EM loop or not
+ //double m_convergence_threshold; ///< convergence threshold
+ //size_t m_max_iterations; ///< maximum number of EM iterations
+
+
/**
* @brief The initialization method
* Check that there is no duplicated means during the random initialization