Added more tests with blitz arrays

bob/learn/em/MAP_gmm_trainer.cpp

@@ -185,7 +185,7 @@ PyObject* PyBobLearnEMMAPGMMTrainer_getRelevanceFactor(PyBobLearnEMMAPGMMTrainer
 int PyBobLearnEMMAPGMMTrainer_setRelevanceFactor(PyBobLearnEMMAPGMMTrainerObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if(!PyNumber_Check(value)){
+  if(!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects a double", Py_TYPE(self)->tp_name, relevance_factor.name());
     return -1;
   }
@@ -211,7 +211,7 @@ PyObject* PyBobLearnEMMAPGMMTrainer_getAlpha(PyBobLearnEMMAPGMMTrainerObject* se
 int PyBobLearnEMMAPGMMTrainer_setAlpha(PyBobLearnEMMAPGMMTrainerObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if(!PyNumber_Check(value)){
+  if(!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects a double", Py_TYPE(self)->tp_name, alpha.name());
     return -1;
   }

bob/learn/em/cpp/MAP_GMMTrainer.cpp

@@ -75,6 +75,10 @@ void bob::learn::em::MAP_GMMTrainer::mStep(bob::learn::em::GMMMachine& gmm)
 {
   // Read options and variables
   double n_gaussians = gmm.getNGaussians();
+
+  //Checking if it is necessary to resize the cache
+  if((size_t)m_cache_alpha.extent(0) != n_gaussians)
+    initialize(gmm); //If it is different for some reason, there is no way, you have to initialize  
 
   // Check that the prior GMM has been specified
   if (!m_prior_gmm)

bob/learn/em/cpp/ML_GMMTrainer.cpp

@@ -41,6 +41,10 @@ void bob::learn::em::ML_GMMTrainer::mStep(bob::learn::em::GMMMachine& gmm)
   // Read options and variables
   const size_t n_gaussians = gmm.getNGaussians();
 
+ //Checking if it is necessary to resize the cache
+ if((size_t)m_cache_ss_n_thresholded.extent(0) != n_gaussians)
+   initialize(gmm); //If it is different for some reason, there is no way, you have to initialize
+
   // - Update weights if requested
   //   Equation 9.26 of Bishop, "Pattern recognition and machine learning", 2006
   if (m_gmm_base_trainer.getUpdateWeights()) {
@@ -101,12 +105,3 @@ bool bob::learn::em::ML_GMMTrainer::operator!=
 {
   return !(this->operator==(other));
 }
-
-/*
-bool bob::learn::em::ML_GMMTrainer::is_similar_to
-  (const bob::learn::em::ML_GMMTrainer &other, const double r_epsilon,
-   const double a_epsilon) const
-{
-  return m_gmm_base_trainer.is_similar_to(other, r_epsilon, a_epsilon);
-}
-*/

bob/learn/em/gaussian.cpp

@@ -105,7 +105,7 @@ static int PyBobLearnEMGaussian_init(PyBobLearnEMGaussianObject* self, PyObject*
   }
 
   /**If the constructor input is a number**/
-  if (PyNumber_Check(arg)) 
+  if (PyBob_NumberCheck(arg)) 
     return PyBobLearnEMGaussian_init_number(self, args, kwargs);
   /**If the constructor input is Gaussian object**/
   else if (PyBobLearnEMGaussian_Check(arg))

bob/learn/em/gmm_stats.cpp

@@ -350,7 +350,7 @@ PyObject* PyBobLearnEMGMMStats_getLog_likelihood(PyBobLearnEMGMMStatsObject* sel
 int PyBobLearnEMGMMStats_setLog_likelihood(PyBobLearnEMGMMStatsObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if (!PyNumber_Check(value)){
+  if (!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects an double", Py_TYPE(self)->tp_name, t.name());
     return -1;
   }

bob/learn/em/ivector_machine.cpp

@@ -270,7 +270,7 @@ PyObject* PyBobLearnEMIVectorMachine_getVarianceThreshold(PyBobLearnEMIVectorMac
 int PyBobLearnEMIVectorMachine_setVarianceThreshold(PyBobLearnEMIVectorMachineObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if (!PyNumber_Check(value)){
+  if (!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects an double", Py_TYPE(self)->tp_name, variance_threshold.name());
     return -1;
   }

bob/learn/em/kmeans_trainer.cpp

@@ -49,7 +49,7 @@ static auto KMeansTrainer_doc = bob::extension::ClassDoc(
   .add_prototype("other","")
   .add_prototype("","")
 
-  .add_parameter("initialization_method", "str", "The initialization method of the means")
+  .add_parameter("initialization_method", "str", "The initialization method of the means.\nPossible values are: 'RANDOM', 'RANDOM_NO_DUPLICATE', 'KMEANS_PLUS_PLUS' ")
   .add_parameter("other", ":py:class:`bob.learn.em.KMeansTrainer`", "A KMeansTrainer object to be copied.")
 
 );
@@ -162,7 +162,10 @@ static auto initialization_method = bob::extension::VariableDoc(
   "initialization_method",
   "str",
   "Initialization method.",
-  ""
+  "Possible values:\n"
+  " `RANDOM`: Random initialization \n\n"
+  " `RANDOM_NO_DUPLICATE`: Random initialization without repetition \n\n"
+  " `KMEANS_PLUS_PLUS`: Apply the kmeans++ initialization http://en.wikipedia.org/wiki/K-means%2B%2B  \n\n"
 );
 PyObject* PyBobLearnEMKMeansTrainer_getInitializationMethod(PyBobLearnEMKMeansTrainerObject* self, void*) {
   BOB_TRY
@@ -254,7 +257,7 @@ PyObject* PyBobLearnEMKMeansTrainer_getAverageMinDistance(PyBobLearnEMKMeansTrai
 int PyBobLearnEMKMeansTrainer_setAverageMinDistance(PyBobLearnEMKMeansTrainerObject* self, PyObject* value, void*) {
   BOB_TRY
 
-  if (!PyNumber_Check(value)){
+  if (!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects an double", Py_TYPE(self)->tp_name, average_min_distance.name());
     return -1;
   }

bob/learn/em/plda_base.cpp

@@ -397,7 +397,7 @@ static PyObject* PyBobLearnEMPLDABase_getVarianceThreshold(PyBobLearnEMPLDABaseO
 int PyBobLearnEMPLDABase_setVarianceThreshold(PyBobLearnEMPLDABaseObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if (!PyNumber_Check(value)){
+  if (!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects an float", Py_TYPE(self)->tp_name, variance_threshold.name());
     return -1;
   }

bob/learn/em/plda_machine.cpp

@@ -220,7 +220,7 @@ static PyObject* PyBobLearnEMPLDAMachine_getWSumXitBetaXi(PyBobLearnEMPLDAMachin
 int PyBobLearnEMPLDAMachine_setWSumXitBetaXi(PyBobLearnEMPLDAMachineObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if (!PyNumber_Check(value)){
+  if (!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects an float", Py_TYPE(self)->tp_name, w_sum_xit_beta_xi.name());
     return -1;
   }
@@ -312,7 +312,7 @@ static PyObject* PyBobLearnEMPLDAMachine_getLogLikelihood(PyBobLearnEMPLDAMachin
 int PyBobLearnEMPLDAMachine_setLogLikelihood(PyBobLearnEMPLDAMachineObject* self, PyObject* value, void*){
   BOB_TRY
 
-  if (!PyNumber_Check(value)){
+  if (!PyBob_NumberCheck(value)){
     PyErr_Format(PyExc_RuntimeError, "%s %s expects an double", Py_TYPE(self)->tp_name, log_likelihood.name());
     return -1;
   }

version.txt

@@ -1 +1 @@
-2.0.0b3
+2.0.0b6