diff --git a/bob/learn/misc/MAP_gmm_trainer.cpp b/bob/learn/misc/MAP_gmm_trainer.cpp
index 66b3f04eee3ad02261476d13bc9e61316fa32574..48ce97c8b2d25cae94c868d839a44d221544517e 100644
--- a/bob/learn/misc/MAP_gmm_trainer.cpp
+++ b/bob/learn/misc/MAP_gmm_trainer.cpp
@@ -13,7 +13,6 @@
/************ Constructor Section *********************************/
/******************************************************************/
-static inline bool f(PyObject* o){return o != 0 && PyObject_IsTrue(o) > 0;} /* converts PyObject to bool and returns false if object is NULL */
static auto MAP_GMMTrainer_doc = bob::extension::ClassDoc(
BOB_EXT_MODULE_PREFIX ".MAP_GMMTrainer",
@@ -27,8 +26,8 @@ static auto MAP_GMMTrainer_doc = bob::extension::ClassDoc(
)
- //.add_prototype("gmm_base_trainer,prior_gmm,[reynolds_adaptation],[relevance_factor],[alpha]","")
- .add_prototype("gmm_base_trainer, prior_gmm, reynolds_adaptation, [relevance_factor], [alpha]","")
+ .add_prototype("gmm_base_trainer,prior_gmm,relevance_factor","")
+ .add_prototype("gmm_base_trainer,prior_gmm,alpha","")
.add_prototype("other","")
.add_prototype("","")
@@ -43,7 +42,7 @@ static auto MAP_GMMTrainer_doc = bob::extension::ClassDoc(
static int PyBobLearnMiscMAPGMMTrainer_init_copy(PyBobLearnMiscMAPGMMTrainerObject* self, PyObject* args, PyObject* kwargs) {
- char** kwlist = MAP_GMMTrainer_doc.kwlist(1);
+ char** kwlist = MAP_GMMTrainer_doc.kwlist(2);
PyBobLearnMiscMAPGMMTrainerObject* o;
if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!", kwlist, &PyBobLearnMiscMAPGMMTrainer_Type, &o)){
MAP_GMMTrainer_doc.print_usage();
@@ -57,25 +56,46 @@ static int PyBobLearnMiscMAPGMMTrainer_init_copy(PyBobLearnMiscMAPGMMTrainerObje
static int PyBobLearnMiscMAPGMMTrainer_init_base_trainer(PyBobLearnMiscMAPGMMTrainerObject* self, PyObject* args, PyObject* kwargs) {
- char** kwlist = MAP_GMMTrainer_doc.kwlist(0);
+ char** kwlist1 = MAP_GMMTrainer_doc.kwlist(0);
+ char** kwlist2 = MAP_GMMTrainer_doc.kwlist(1);
PyBobLearnMiscGMMBaseTrainerObject* gmm_base_trainer;
PyBobLearnMiscGMMMachineObject* gmm_machine;
- PyObject* reynolds_adaptation = 0;
+ bool reynolds_adaptation = false;
double alpha = 0.5;
double relevance_factor = 4.0;
+ double aux = 0;
+ PyObject* keyword_relevance_factor = Py_BuildValue("s", kwlist1[2]);
+ PyObject* keyword_alpha = Py_BuildValue("s", kwlist2[2]);
- if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O!|O!dd", kwlist, &PyBobLearnMiscGMMBaseTrainer_Type, &gmm_base_trainer,
+ //Here we have to select which keyword argument to read
+ if (kwargs && PyDict_Contains(kwargs, keyword_relevance_factor) && (PyArg_ParseTupleAndKeywords(args, kwargs, "O!O!|d", kwlist1,
+ &PyBobLearnMiscGMMBaseTrainer_Type, &gmm_base_trainer,
&PyBobLearnMiscGMMMachine_Type, &gmm_machine,
- &PyBool_Type, &reynolds_adaptation,
- &relevance_factor, &alpha )){
-
+ &aux)))
+ reynolds_adaptation = true;
+ else if (kwargs && PyDict_Contains(kwargs, keyword_alpha) && (PyArg_ParseTupleAndKeywords(args, kwargs, "O!O!|d", kwlist2,
+ &PyBobLearnMiscGMMBaseTrainer_Type, &gmm_base_trainer,
+ &PyBobLearnMiscGMMMachine_Type, &gmm_machine,
+ &aux)))
+ reynolds_adaptation = false;
+ else{
+ std::cout << PyDict_Contains(kwargs, keyword_alpha) << std::endl;
+ PyErr_Format(PyExc_RuntimeError, "%s. The third argument must be a keyword argument.", Py_TYPE(self)->tp_name);
MAP_GMMTrainer_doc.print_usage();
return -1;
}
+
+
+
+ if (reynolds_adaptation)
+ relevance_factor = aux;
+ else
+ alpha = aux;
- self->cxx.reset(new bob::learn::misc::MAP_GMMTrainer(gmm_base_trainer->cxx, gmm_machine->cxx, f(reynolds_adaptation),relevance_factor, alpha));
+
+ self->cxx.reset(new bob::learn::misc::MAP_GMMTrainer(gmm_base_trainer->cxx, gmm_machine->cxx, reynolds_adaptation,relevance_factor, alpha));
return 0;
}
@@ -172,31 +192,6 @@ int PyBobLearnMiscMAPGMMTrainer_setGMMBaseTrainer(PyBobLearnMiscMAPGMMTrainerObj
}
-/***** reynolds_adaptation *****/
-static auto reynolds_adaptation = bob::extension::VariableDoc(
- "reynolds_adaptation",
- "bool",
- "Will use the Reynolds adaptation factor? See Eq (14) from [Reynolds2000]_",
- ""
-);
-PyObject* PyBobLearnMiscMAPGMMTrainer_getReynoldsAdaptation(PyBobLearnMiscMAPGMMTrainerObject* self, void*){
- BOB_TRY
- return Py_BuildValue("O",self->cxx->getReynoldsAdaptation()?Py_True:Py_False);
- BOB_CATCH_MEMBER("reynolds_adaptation could not be read", 0)
-}
-int PyBobLearnMiscMAPGMMTrainer_setReynoldsAdaptation(PyBobLearnMiscMAPGMMTrainerObject* self, PyObject* value, void*){
- BOB_TRY
-
- if(!PyBool_Check(value)){
- PyErr_Format(PyExc_RuntimeError, "%s %s expects a boolean", Py_TYPE(self)->tp_name, reynolds_adaptation.name());
- return -1;
- }
-
- self->cxx->setReynoldsAdaptation(f(value));
- return 0;
- BOB_CATCH_MEMBER("reynolds_adaptation could not be set", 0)
-}
-
/***** relevance_factor *****/
static auto relevance_factor = bob::extension::VariableDoc(
@@ -218,7 +213,7 @@ int PyBobLearnMiscMAPGMMTrainer_setRelevanceFactor(PyBobLearnMiscMAPGMMTrainerOb
return -1;
}
- self->cxx->setRelevanceFactor(f(value));
+ self->cxx->setRelevanceFactor(PyFloat_AS_DOUBLE(value));
return 0;
BOB_CATCH_MEMBER("relevance_factor could not be set", 0)
}
@@ -244,7 +239,7 @@ int PyBobLearnMiscMAPGMMTrainer_setAlpha(PyBobLearnMiscMAPGMMTrainerObject* self
return -1;
}
- self->cxx->setAlpha(f(value));
+ self->cxx->setAlpha(PyFloat_AS_DOUBLE(value));
return 0;
BOB_CATCH_MEMBER("alpha could not be set", 0)
}
@@ -259,13 +254,6 @@ static PyGetSetDef PyBobLearnMiscMAPGMMTrainer_getseters[] = {
gmm_base_trainer.doc(),
0
},
- {
- reynolds_adaptation.name(),
- (getter)PyBobLearnMiscMAPGMMTrainer_getReynoldsAdaptation,
- (setter)PyBobLearnMiscMAPGMMTrainer_setReynoldsAdaptation,
- reynolds_adaptation.doc(),
- 0
- },
{
alpha.name(),
(getter)PyBobLearnMiscMAPGMMTrainer_getAlpha,
diff --git a/bob/learn/misc/__MAP_gmm_trainer__.py b/bob/learn/misc/__MAP_gmm_trainer__.py
index a41855bcf5650012ab60ea11ae79393c5d541b4b..18f215dcb9c1a2331b09ed27812f5f3ef739241b 100644
--- a/bob/learn/misc/__MAP_gmm_trainer__.py
+++ b/bob/learn/misc/__MAP_gmm_trainer__.py
@@ -11,7 +11,7 @@ import numpy
# define the class
class MAP_GMMTrainer(_MAP_GMMTrainer):
- def __init__(self, gmm_base_trainer, prior_gmm, convergence_threshold=0.001, max_iterations=10, converge_by_likelihood=True, reynolds_adaptation=False, relevance_factor=4., alpha=0.5):
+ def __init__(self, gmm_base_trainer, prior_gmm, convergence_threshold=0.001, max_iterations=10, converge_by_likelihood=True, **kwargs):
"""
:py:class:`bob.learn.misc.MAP_GMMTrainer` constructor
@@ -26,21 +26,27 @@ class MAP_GMMTrainer(_MAP_GMMTrainer):
Number of maximum iterations
converge_by_likelihood
Tells whether we compute log_likelihood as a convergence criteria, or not
- reynolds_adaptation
- Will use the Reynolds adaptation procedure? See Eq (14) from [Reynolds2000]_
- relevance_factor
- If set the :py:class:`bob.learn.misc.MAP_GMMTrainer.reynolds_adaptation` parameters, will apply the Reynolds Adaptation procedure. See Eq (14) from [Reynolds2000]_
alpha
Set directly the alpha parameter (Eq (14) from [Reynolds2000]_), ignoring zeroth order statistics as a weighting factor.
+ relevance_factor
+ If set the :py:class:`bob.learn.misc.MAP_GMMTrainer.reynolds_adaptation` parameters, will apply the Reynolds Adaptation procedure. See Eq (14) from [Reynolds2000]_
"""
- _MAP_GMMTrainer.__init__(self, gmm_base_trainer, prior_gmm, reynolds_adaptation=reynolds_adaptation, relevance_factor=relevance_factor, alpha=alpha)
+ if kwargs.get('alpha')!=None:
+ alpha = kwargs.get('alpha')
+ _MAP_GMMTrainer.__init__(self, gmm_base_trainer, prior_gmm, alpha=alpha)
+ else:
+ relevance_factor = kwargs.get('relevance_factor')
+ _MAP_GMMTrainer.__init__(self, gmm_base_trainer, prior_gmm, relevance_factor=relevance_factor)
self.convergence_threshold = convergence_threshold
self.max_iterations = max_iterations
self.converge_by_likelihood = converge_by_likelihood
+
+
+
def train(self, gmm_machine, data):
"""
Train the :py:class:bob.learn.misc.GMMMachine using data
diff --git a/bob/learn/misc/__kmeans_trainer__.py b/bob/learn/misc/__kmeans_trainer__.py
index c4bab14e1780e099a90ae85cb4ab8ff309de80ad..afe4949366f936138d11ada07f87c0c3355b3a87 100644
--- a/bob/learn/misc/__kmeans_trainer__.py
+++ b/bob/learn/misc/__kmeans_trainer__.py
@@ -13,7 +13,7 @@ class KMeansTrainer (_KMeansTrainer):
def __init__(self, initialization_method="RANDOM", convergence_threshold=0.001, max_iterations=10, converge_by_average_min_distance=True):
"""
- :py:class:bob.learn.misc.KMeansTrainer constructor
+ :py:class:`bob.learn.misc.KMeansTrainer` constructor
Keyword Parameters:
initialization_method
@@ -78,4 +78,4 @@ class KMeansTrainer (_KMeansTrainer):
# copy the documentation from the base class
-__doc__ = _KMeansTrainer.__doc__
\ No newline at end of file
+__doc__ = _KMeansTrainer.__doc__
diff --git a/bob/learn/misc/test_em.py b/bob/learn/misc/test_em.py
index 31e510e887ddb14930e04d768b7d31a8bcf9302c..dffb86a9951921ff78c8597e10d31895d32599c5 100644
--- a/bob/learn/misc/test_em.py
+++ b/bob/learn/misc/test_em.py
@@ -115,7 +115,7 @@ def test_gmm_MAP_1():
gmm = GMMMachine(bob.io.base.HDF5File(datafile("gmm_ML.hdf5", __name__)))
gmmprior = GMMMachine(bob.io.base.HDF5File(datafile("gmm_ML.hdf5", __name__)))
- map_gmmtrainer = MAP_GMMTrainer(GMMBaseTrainer(True, False, False),gmmprior, reynolds_adaptation=False, relevance_factor=16)
+ map_gmmtrainer = MAP_GMMTrainer(GMMBaseTrainer(True, False, False),gmmprior, relevance_factor=4.)
#map_gmmtrainer.set_prior_gmm(gmmprior)
map_gmmtrainer.train(gmm, ar)
@@ -142,7 +142,7 @@ def test_gmm_MAP_2():
gmm.variances = variances
gmm.weights = weights
- map_adapt = MAP_GMMTrainer(GMMBaseTrainer(True, False, False, mean_var_update_responsibilities_threshold=0.),gmm, alpha=4., relevance_factor=4., reynolds_adaptation=True)
+ map_adapt = MAP_GMMTrainer(GMMBaseTrainer(True, False, False, mean_var_update_responsibilities_threshold=0.),gmm, relevance_factor=4.)
#map_adapt.set_prior_gmm(gmm)
gmm_adapted = GMMMachine(2,50)
@@ -186,7 +186,7 @@ def test_gmm_MAP_3():
max_iter_gmm = 1
accuracy = 0.00001
map_factor = 0.5
- map_gmmtrainer = MAP_GMMTrainer(GMMBaseTrainer(True, False, False, prior), prior_gmm, alpha=map_factor, reynolds_adaptation=False)
+ map_gmmtrainer = MAP_GMMTrainer(GMMBaseTrainer(True, False, False, prior), prior_gmm, alpha=map_factor)
map_gmmtrainer.max_iterations = max_iter_gmm
map_gmmtrainer.convergence_threshold = accuracy