Fixed the documentation and added the random number generator as a parameter in the general trainer

bob/learn/em/empca_trainer.cpp

@@ -14,7 +14,7 @@
 /******************************************************************/
 
 static auto EMPCATrainer_doc = bob::extension::ClassDoc(
-  BOB_EXT_MODULE_PREFIX "._EMPCATrainer",
+  BOB_EXT_MODULE_PREFIX ".EMPCATrainer",
   ""
 
 ).add_constructor(
@@ -338,6 +338,6 @@ bool init_BobLearnEMEMPCATrainer(PyObject* module)
 
   // add the type to the module
   Py_INCREF(&PyBobLearnEMEMPCATrainer_Type);
-  return PyModule_AddObject(module, "_EMPCATrainer", (PyObject*)&PyBobLearnEMEMPCATrainer_Type) >= 0;
+  return PyModule_AddObject(module, "EMPCATrainer", (PyObject*)&PyBobLearnEMEMPCATrainer_Type) >= 0;
 }
 

bob/learn/em/plda_trainer.cpp

@@ -287,7 +287,7 @@ static auto init_g_method = bob::extension::VariableDoc(
   "init_g_method",
   "str",
   "The method used for the initialization of :math:`$G$`.",
-  "Possible values are: ('RANDOM_G', 'BETWEEN_SCATTER')"
+  "Possible values are: ('RANDOM_G', 'WITHIN_SCATTER')"
 );
 PyObject* PyBobLearnEMPLDATrainer_getGMethod(PyBobLearnEMPLDATrainerObject* self, void*) {
   BOB_TRY

bob/learn/em/train.py

@@ -7,27 +7,32 @@
 import numpy
 import bob.learn.em
 
-def train(trainer, machine, data, max_iterations = 50, convergence_threshold=None, initialize=True):
+def train(trainer, machine, data, max_iterations = 50, convergence_threshold=None, initialize=True, rng=None):
   """
   Trains a machine given a trainer and the proper data
 
   **Parameters**:
-    trainer
+    trainer : one of :py:class:`KMeansTrainer`, :py:class:`MAP_GMMTrainer`, :py:class:`ML_GMMTrainer`, :py:class:`ISVTrainer`, :py:class:`IVectorTrainer`, :py:class:`PLDATrainer`, :py:class:`EMPCATrainer`
       A trainer mechanism
-    machine
+    machine : one of :py:class:`KMeansMachine`, :py:class:`GMMMachine`, :py:class:`ISVBase`, :py:class:`IVectorMachine`, :py:class:`PLDAMachine`, :py:class:`bob.learn.linear.Machine`
       A container machine
-    data 
+    data : array_like <float, 2D>
       The data to be trained
-    max_iterations
+    max_iterations : int
       The maximum number of iterations to train a machine
-    convergence_threshold
+    convergence_threshold : float
       The convergence threshold to train a machine. If None, the training procedure will stop with the iterations criteria
-    initialize
+    initialize : bool
       If True, runs the initialization procedure
+    rng :  :py:class:`bob.core.random.mt19937`
+      The Mersenne Twister mt19937 random generator used for the initialization of subspaces/arrays before the EM loop
   """
   #Initialization
   if initialize:
-    trainer.initialize(machine, data)
+    if rng is not None:
+      trainer.initialize(machine, data, rng)
+    else:
+      trainer.initialize(machine, data)
 
   trainer.eStep(machine, data)  
   average_output          = 0

doc/py_api.rst

@@ -12,9 +12,11 @@ This section includes information for using the pure Python API of
 Classes
 -------
 
+
+Trainers
+........
+
 .. autosummary::
-  Trainers
-  --------
   
   bob.learn.em.KMeansTrainer
   bob.learn.em.ML_GMMTrainer
@@ -23,9 +25,12 @@ Classes
   bob.learn.em.JFATrainer  
   bob.learn.em.IVectorTrainer
   bob.learn.em.PLDATrainer
+  bob.learn.em.EMPCATrainer
   
-  Machines
-  --------
+Machines
+........
+
+.. autosummary::  
   
   bob.learn.em.KMeansMachine
   bob.learn.em.Gaussian