Compare revisions

akomaty@idiap.ch · Amir MOHAMMADI · Tiago Pereira · Amir MOHAMMADI · Amir MOHAMMADI · Tiago Pereira
--- a/.gitignore
+++ b/.gitignore
@@ -15,5 +15,6 @@ dist
 .nfs*
 .gdb_history
 build
+.DS_Store
 *.egg
 src/
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -142,31 +142,6 @@ deploy_linux_27:
    - conda-linux
-# Linux + Python 3.4: Builds and tests
-build_linux_34:
-  <<: *build_job
-  variables: &linux_34_build_variables
-    PYTHON_VERSION: "3.4"
-  tags:
-    - conda-linux
-test_linux_34:
-  <<: *test_job
-  variables: *linux_34_build_variables
-  dependencies:
-    - build_linux_34
-  tags:
-    - conda-linux
-wheels_linux_34:
-  <<: *wheels_job
-  variables: *linux_34_build_variables
-  dependencies:
-    - build_linux_34
-  tags:
-    - conda-linux
 # Linux + Python 3.5: Builds, tests and uploads wheel
 build_linux_35:
  <<: *build_job
@@ -200,6 +175,31 @@ docs_linux_35:
    - conda-linux
+# Linux + Python 3.6: Builds and tests
+build_linux_36:
+  <<: *build_job
+  variables: &linux_36_build_variables
+    PYTHON_VERSION: "3.6"
+  tags:
+    - conda-linux
+test_linux_36:
+  <<: *test_job
+  variables: *linux_36_build_variables
+  dependencies:
+    - build_linux_36
+  tags:
+    - conda-linux
+wheels_linux_36:
+  <<: *wheels_job
+  variables: *linux_36_build_variables
+  dependencies:
+    - build_linux_36
+  tags:
+    - conda-linux
 # Mac OSX + Python 2.7: Builds and tests
 build_macosx_27:
  <<: *build_job
@@ -225,51 +225,51 @@ wheels_macosx_27:
    - conda-macosx
-# Mac OSX + Python 3.4: Builds and tests
+# Mac OSX + Python 3.5: Builds and tests
-build_macosx_34:
+build_macosx_35:
  <<: *build_job
-  variables: &macosx_34_build_variables
+  variables: &macosx_35_build_variables
-    PYTHON_VERSION: "3.4"
+    PYTHON_VERSION: "3.5"
  tags:
    - conda-macosx
-test_macosx_34:
+test_macosx_35:
  <<: *test_job
-  variables: *macosx_34_build_variables
+  variables: *macosx_35_build_variables
  dependencies:
-    - build_macosx_34
+    - build_macosx_35
  tags:
    - conda-macosx
-wheels_macosx_34:
+wheels_macosx_35:
  <<: *wheels_job
-  variables: *macosx_34_build_variables
+  variables: *macosx_35_build_variables
  dependencies:
-    - build_macosx_34
+    - build_macosx_35
  tags:
    - conda-macosx
-# Mac OSX + Python 3.5: Builds and tests
+# Mac OSX + Python 3.6: Builds and tests
-build_macosx_35:
+build_macosx_36:
  <<: *build_job
-  variables: &macosx_35_build_variables
+  variables: &macosx_36_build_variables
-    PYTHON_VERSION: "3.5"
+    PYTHON_VERSION: "3.6"
  tags:
    - conda-macosx
-test_macosx_35:
+test_macosx_36:
  <<: *test_job
-  variables: *macosx_35_build_variables
+  variables: *macosx_36_build_variables
  dependencies:
-    - build_macosx_35
+    - build_macosx_36
  tags:
    - conda-macosx
-wheels_macosx_35:
+wheels_macosx_36:
  <<: *wheels_job
-  variables: *macosx_35_build_variables
+  variables: *macosx_36_build_variables
  dependencies:
-    - build_macosx_35
+    - build_macosx_36
  tags:
    - conda-macosx
--- a/MANIFEST.in
+++ b/MANIFEST.in
-include LICENSE README.rst bootstrap-buildout.py buildout.cfg develop.cfg requirements.txt version.txt
+include LICENSE README.rst buildout.cfg develop.cfg requirements.txt version.txt
 recursive-include doc conf.py *.rst
 recursive-include bob/learn/em *.cpp *.h
 recursive-include bob/learn/em/data *.*
--- a/README.rst
+++ b/README.rst
@@ -7,6 +7,8 @@
   :target: https://www.idiap.ch/software/bob/docs/latest/bob/bob.learn.em/master/index.html
 .. image:: https://gitlab.idiap.ch/bob/bob.learn.em/badges/master/build.svg
   :target: https://gitlab.idiap.ch/bob/bob.learn.em/commits/master
+.. image:: https://gitlab.idiap.ch/bob/bob.learn.em/badges/master/coverage.svg
+   :target: https://gitlab.idiap.ch/bob/bob.learn.em/commits/master
 .. image:: https://img.shields.io/badge/gitlab-project-0000c0.svg
   :target: https://gitlab.idiap.ch/bob/bob.learn.em
 .. image:: http://img.shields.io/pypi/v/bob.learn.em.svg
@@ -47,11 +49,10 @@ The package includes the machine definition per se and a selection of different
 Installation
 ------------
-Follow our `installation`_ instructions. Then, using the Python interpreter
+Complete Bob's `installation`_ instructions. Then, to install this pacakge,
-provided by the distribution, bootstrap and buildout this package::
+run::
-  $ python bootstrap-buildout.py
+  $ conda install bob.learn.em
-  $ ./bin/buildout
 Contact

--- a/bob/learn/em/cpp/IVectorMachine.cpp
+++ b/bob/learn/em/cpp/IVectorMachine.cpp
@@ -206,7 +206,7 @@ void bob::learn::em::IVectorMachine::resizeTmp()
 void bob::learn::em::IVectorMachine::forward(const bob::learn::em::GMMStats& gs,
  blitz::Array<double,1>& ivector) const
 {
-  bob::core::array::assertSameDimensionLength(ivector.extent(0), (int)m_rt);  
+  bob::core::array::assertSameDimensionLength(ivector.extent(0), (int)m_rt);
  forward_(gs, ivector);
 }
@@ -246,6 +246,5 @@ void bob::learn::em::IVectorMachine::forward_(const bob::learn::em::GMMStats& gs
  computeTtSigmaInvFnorm(gs, m_tmp_t1);
  // Solves m_tmp_tt.ivector = m_tmp_t1
-  bob::math::linsolve(m_tmp_tt, ivector, m_tmp_t1);
+  bob::math::linsolve(m_tmp_tt, m_tmp_t1, ivector);
 }
--- a/bob/learn/em/cpp/IVectorTrainer.cpp
+++ b/bob/learn/em/cpp/IVectorTrainer.cpp
@@ -170,7 +170,7 @@ void bob::learn::em::IVectorTrainer::mStep(
    if (blitz::all(acc_Nij_wij2_c == 0)) // TODO
      Tt_c = 0;
    else
-      bob::math::linsolve(tacc_Nij_wij2_c, Tt_c, tacc_Fnormij_wij_c);
+      bob::math::linsolve(tacc_Nij_wij2_c, tacc_Fnormij_wij_c, Tt_c);
    if (m_update_sigma)
    {
      blitz::Array<double,1> sigma_c = sigma(blitz::Range(c*D,(c+1)*D-1));

--- a/bob/learn/em/gmm_machine.cpp
+++ b/bob/learn/em/gmm_machine.cpp
@@ -15,7 +15,8 @@
 static auto GMMMachine_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".GMMMachine",
-  "This class implements a multivariate diagonal Gaussian distribution.",
+  "This class implements the statistical model for multivariate diagonal mixture Gaussian distribution (GMM). "
+  "A GMM is defined as :math:`\\sum_{c=0}^{C} \\omega_c \\mathcal{N}(x | \\mu_c, \\sigma_c)`, where :math:`C` is the number of Gaussian components :math:`\\mu_c`, :math:`\\sigma_c` and :math:`\\omega_c` are respectively the the mean, variance and the weight of each gaussian component :math:`c`.",
  "See Section 2.3.9 of Bishop, \"Pattern recognition and machine learning\", 2006"
 ).add_constructor(
  bob::extension::FunctionDoc(
@@ -744,7 +745,7 @@ static PyObject* PyBobLearnEMGMMMachine_loglikelihood_(PyBobLearnEMGMMMachineObj
 /*** acc_statistics ***/
 static auto acc_statistics = bob::extension::FunctionDoc(
  "acc_statistics",
-  "Accumulate the GMM statistics for this sample(s). Inputs are checked.",
+  "Accumulate the GMM statistics (:py:class:`bob.learn.em.GMMStats`) for this sample(s). Inputs are checked.",
  "",
  true
 )
@@ -780,7 +781,7 @@ static PyObject* PyBobLearnEMGMMMachine_accStatistics(PyBobLearnEMGMMMachineObje
 /*** acc_statistics_ ***/
 static auto acc_statistics_ = bob::extension::FunctionDoc(
  "acc_statistics_",
-  "Accumulate the GMM statistics for this sample(s). Inputs are NOT checked.",
+  "Accumulate the GMM statistics (:py:class:`bob.learn.em.GMMStats`) for this sample(s). Inputs are NOT checked.",
  "",
  true
 )
@@ -853,7 +854,7 @@ static PyObject* PyBobLearnEMGMMMachine_setVarianceThresholds_method(PyBobLearnE
 /*** get_gaussian ***/
 static auto get_gaussian = bob::extension::FunctionDoc(
  "get_gaussian",
-  "Get the specified Gaussian component.",
+  "Get the specified Gaussian (:py:class:`bob.learn.em.Gaussian`) component.",
  ".. note:: An exception is thrown if i is out of range.",
  true
 )

--- a/bob/learn/em/include/bob.learn.em/KMeansTrainer.h
+++ b/bob/learn/em/include/bob.learn.em/KMeansTrainer.h
@@ -40,7 +40,7 @@ class KMeansTrainer
    /**
     * @brief Constructor
     */
-    KMeansTrainer(InitializationMethod=RANDOM);
+    KMeansTrainer(InitializationMethod=RANDOM_NO_DUPLICATE);
    /**
     * @brief Virtualize destructor

--- a/bob/learn/em/isv_base.cpp
+++ b/bob/learn/em/isv_base.cpp
@@ -184,7 +184,7 @@ static auto supervector_length = bob::extension::VariableDoc(
  "int",
  "Returns the supervector length.",
  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality"
-  "WARNING An exception is thrown if no Universal Background Model has been set yet."
+  "An exception is thrown if no Universal Background Model has been set yet."
  ""
 );
 PyObject* PyBobLearnEMISVBase_getSupervectorLength(PyBobLearnEMISVBaseObject* self, void*) {

--- a/bob/learn/em/isv_machine.cpp
+++ b/bob/learn/em/isv_machine.cpp
@@ -69,15 +69,15 @@ static int PyBobLearnEMISVMachine_init_hdf5(PyBobLearnEMISVMachineObject* self,
 static int PyBobLearnEMISVMachine_init_isvbase(PyBobLearnEMISVMachineObject* self, PyObject* args, PyObject* kwargs) {
  char** kwlist = ISVMachine_doc.kwlist(0);
  PyBobLearnEMISVBaseObject* isv_base;
-  //Here we have to select which keyword argument to read  
+  //Here we have to select which keyword argument to read
  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!", kwlist, &PyBobLearnEMISVBase_Type, &isv_base)){
    ISVMachine_doc.print_usage();
    return -1;
  }
  self->cxx.reset(new bob::learn::em::ISVMachine(isv_base->cxx));
  return 0;
 }
@@ -115,7 +115,7 @@ static int PyBobLearnEMISVMachine_init(PyBobLearnEMISVMachineObject* self, PyObj
    ISVMachine_doc.print_usage();
    return -1;
  }
  BOB_CATCH_MEMBER("cannot create ISVMachine", -1)
  return 0;
 }
@@ -173,8 +173,8 @@ static auto supervector_length = bob::extension::VariableDoc(
  "int",
  "Returns the supervector length.",
-  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality"
+  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality. "
-  "@warning An exception is thrown if no Universal Background Model has been set yet."
+  "An exception is thrown if no Universal Background Model has been set yet."
  ""
 );
 PyObject* PyBobLearnEMISVMachine_getSupervectorLength(PyBobLearnEMISVMachineObject* self, void*) {
@@ -203,17 +203,17 @@ int PyBobLearnEMISVMachine_setZ(PyBobLearnEMISVMachineObject* self, PyObject* va
    return -1;
  }
  auto o_ = make_safe(input);
-  // perform check on the input  
+  // perform check on the input
  if (input->type_num != NPY_FLOAT64){
    PyErr_Format(PyExc_TypeError, "`%s' only supports 64-bit float arrays for input array `%s`", Py_TYPE(self)->tp_name, Z.name());
    return -1;
-  }  
+  }
  if (input->ndim != 1){
    PyErr_Format(PyExc_TypeError, "`%s' only processes 1D arrays of float64 for `%s`", Py_TYPE(self)->tp_name, Z.name());
    return -1;
-  }  
+  }
  if (input->shape[0] != (Py_ssize_t)self->cxx->getZ().extent(0)) {
    PyErr_Format(PyExc_TypeError, "`%s' 1D `input` array should have %" PY_FORMAT_SIZE_T "d, elements, not %" PY_FORMAT_SIZE_T "d for `%s`", Py_TYPE(self)->tp_name, (Py_ssize_t)self->cxx->getZ().extent(0), (Py_ssize_t)input->shape[0], Z.name());
@@ -276,13 +276,13 @@ int PyBobLearnEMISVMachine_setISVBase(PyBobLearnEMISVMachineObject* self, PyObje
  self->cxx->setISVBase(isv_base_o->cxx);
  return 0;
-  BOB_CATCH_MEMBER("isv_base could not be set", -1)  
+  BOB_CATCH_MEMBER("isv_base could not be set", -1)
 }
-static PyGetSetDef PyBobLearnEMISVMachine_getseters[] = { 
+static PyGetSetDef PyBobLearnEMISVMachine_getseters[] = {
  {
   shape.name(),
   (getter)PyBobLearnEMISVMachine_getShape,
@@ -290,7 +290,7 @@ static PyGetSetDef PyBobLearnEMISVMachine_getseters[] = {
   shape.doc(),
   0
  },
  {
   supervector_length.name(),
   (getter)PyBobLearnEMISVMachine_getSupervectorLength,
@@ -298,7 +298,7 @@ static PyGetSetDef PyBobLearnEMISVMachine_getseters[] = {
   supervector_length.doc(),
   0
  },
  {
   isv_base.name(),
   (getter)PyBobLearnEMISVMachine_getISVBase,
@@ -343,9 +343,9 @@ static auto save = bob::extension::FunctionDoc(
 static PyObject* PyBobLearnEMISVMachine_Save(PyBobLearnEMISVMachineObject* self,  PyObject* args, PyObject* kwargs) {
  BOB_TRY
  // get list of arguments
-  char** kwlist = save.kwlist(0);  
+  char** kwlist = save.kwlist(0);
  PyBobIoHDF5FileObject* hdf5;
  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O&", kwlist, PyBobIoHDF5File_Converter, &hdf5)) return 0;
@@ -365,12 +365,12 @@ static auto load = bob::extension::FunctionDoc(
 .add_parameter("hdf5", ":py:class:`bob.io.base.HDF5File`", "An HDF5 file open for reading");
 static PyObject* PyBobLearnEMISVMachine_Load(PyBobLearnEMISVMachineObject* self, PyObject* args, PyObject* kwargs) {
  BOB_TRY
-  char** kwlist = load.kwlist(0);  
+  char** kwlist = load.kwlist(0);
  PyBobIoHDF5FileObject* hdf5;
  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O&", kwlist, PyBobIoHDF5File_Converter, &hdf5)) return 0;
-  auto hdf5_ = make_safe(hdf5);  
+  auto hdf5_ = make_safe(hdf5);
  self->cxx->load(*hdf5->f);
  BOB_CATCH_MEMBER("cannot load the data", 0)
@@ -381,7 +381,7 @@ static PyObject* PyBobLearnEMISVMachine_Load(PyBobLearnEMISVMachineObject* self,
 /*** is_similar_to ***/
 static auto is_similar_to = bob::extension::FunctionDoc(
  "is_similar_to",
  "Compares this ISVMachine with the ``other`` one to be approximately the same.",
  "The optional values ``r_epsilon`` and ``a_epsilon`` refer to the "
  "relative and absolute precision for the ``weights``, ``biases`` "
@@ -406,8 +406,8 @@ static PyObject* PyBobLearnEMISVMachine_IsSimilarTo(PyBobLearnEMISVMachineObject
        &PyBobLearnEMISVMachine_Type, &other,
        &r_epsilon, &a_epsilon)){
-        is_similar_to.print_usage(); 
+        is_similar_to.print_usage();
-        return 0;        
+        return 0;
  }
  if (self->cxx->is_similar_to(*other->cxx, r_epsilon, a_epsilon))
@@ -421,7 +421,7 @@ static PyObject* PyBobLearnEMISVMachine_IsSimilarTo(PyBobLearnEMISVMachineObject
 static auto estimate_x = bob::extension::FunctionDoc(
  "estimate_x",
  "Estimates the session offset x (LPT assumption) given GMM statistics.",
-  "Estimates :math:`x` from the GMM statistics considering the LPT assumption, that is the latent session variable :math:`x` is approximated using the UBM", 
+  "Estimates :math:`x` from the GMM statistics considering the LPT assumption, that is the latent session variable :math:`x` is approximated using the UBM",
  true
 )
 .add_prototype("stats,input")
@@ -435,29 +435,29 @@ static PyObject* PyBobLearnEMISVMachine_estimateX(PyBobLearnEMISVMachineObject*
  PyBobLearnEMGMMStatsObject* stats = 0;
  PyBlitzArrayObject* input           = 0;
-  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O&", kwlist, &PyBobLearnEMGMMStats_Type, &stats, 
+  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O&", kwlist, &PyBobLearnEMGMMStats_Type, &stats,
                                                                 &PyBlitzArray_Converter,&input))
    return 0;
  //protects acquired resources through this scope
  auto input_ = make_safe(input);
-  // perform check on the input  
+  // perform check on the input
  if (input->type_num != NPY_FLOAT64){
    PyErr_Format(PyExc_TypeError, "`%s' only supports 64-bit float arrays for input array `%s`", Py_TYPE(self)->tp_name, estimate_x.name());
    return 0;
-  }  
+  }
  if (input->ndim != 1){
    PyErr_Format(PyExc_TypeError, "`%s' only processes 1D arrays of float64 for `%s`", Py_TYPE(self)->tp_name, estimate_x.name());
    return 0;
-  }  
+  }
  if (input->shape[0] != (Py_ssize_t)self->cxx->getNGaussians()) {
    PyErr_Format(PyExc_TypeError, "`%s' 1D `input` array should have %" PY_FORMAT_SIZE_T "d, elements, not %" PY_FORMAT_SIZE_T "d for `%s`", Py_TYPE(self)->tp_name, self->cxx->getNInputs(), (Py_ssize_t)input->shape[0], estimate_x.name());
    return 0;
  }
  self->cxx->estimateX(*stats->cxx, *PyBlitzArrayCxx_AsBlitz<double,1>(input));
  BOB_CATCH_MEMBER("cannot estimate X", 0)
@@ -469,7 +469,7 @@ static PyObject* PyBobLearnEMISVMachine_estimateX(PyBobLearnEMISVMachineObject*
 static auto estimate_ux = bob::extension::FunctionDoc(
  "estimate_ux",
  "Estimates Ux (LPT assumption) given GMM statistics.",
-  "Estimates Ux from the GMM statistics considering the LPT assumption, that is the latent session variable x is approximated using the UBM.", 
+  "Estimates Ux from the GMM statistics considering the LPT assumption, that is the latent session variable x is approximated using the UBM.",
  true
 )
 .add_prototype("stats,input")
@@ -483,29 +483,29 @@ static PyObject* PyBobLearnEMISVMachine_estimateUx(PyBobLearnEMISVMachineObject*
  PyBobLearnEMGMMStatsObject* stats = 0;
  PyBlitzArrayObject* input           = 0;
-  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O&", kwlist, &PyBobLearnEMGMMStats_Type, &stats, 
+  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O&", kwlist, &PyBobLearnEMGMMStats_Type, &stats,
                                                                 &PyBlitzArray_Converter,&input))
    return 0;
  //protects acquired resources through this scope
  auto input_ = make_safe(input);
-  // perform check on the input  
+  // perform check on the input
  if (input->type_num != NPY_FLOAT64){
    PyErr_Format(PyExc_TypeError, "`%s' only supports 64-bit float arrays for input array `%s`", Py_TYPE(self)->tp_name, estimate_ux.name());
    return 0;
-  }  
+  }
  if (input->ndim != 1){
    PyErr_Format(PyExc_TypeError, "`%s' only processes 1D arrays of float64 for `%s`", Py_TYPE(self)->tp_name, estimate_ux.name());
    return 0;
-  }  
+  }
  if (input->shape[0] != (Py_ssize_t)self->cxx->getNGaussians()*(Py_ssize_t)self->cxx->getNInputs()) {
    PyErr_Format(PyExc_TypeError, "`%s' 1D `input` array should have %" PY_FORMAT_SIZE_T "d, elements, not %" PY_FORMAT_SIZE_T "d for `%s`", Py_TYPE(self)->tp_name, self->cxx->getNInputs()*(Py_ssize_t)self->cxx->getNGaussians(), (Py_ssize_t)input->shape[0], estimate_ux.name());
    return 0;
  }
  self->cxx->estimateUx(*stats->cxx, *PyBlitzArrayCxx_AsBlitz<double,1>(input));
  BOB_CATCH_MEMBER("cannot estimate Ux", 0)
@@ -517,7 +517,7 @@ static PyObject* PyBobLearnEMISVMachine_estimateUx(PyBobLearnEMISVMachineObject*
 static auto forward_ux = bob::extension::FunctionDoc(
  "forward_ux",
  "Computes a score for the given UBM statistics and given the Ux vector",
-  "", 
+  "",
  true
 )
 .add_prototype("stats,ux")
@@ -531,30 +531,30 @@ static PyObject* PyBobLearnEMISVMachine_ForwardUx(PyBobLearnEMISVMachineObject*
  PyBobLearnEMGMMStatsObject* stats = 0;
  PyBlitzArrayObject* ux_input        = 0;
-  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O&", kwlist, &PyBobLearnEMGMMStats_Type, &stats, 
+  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!O&", kwlist, &PyBobLearnEMGMMStats_Type, &stats,
                                                                 &PyBlitzArray_Converter,&ux_input))
    return 0;
  //protects acquired resources through this scope
  auto ux_input_ = make_safe(ux_input);
-  // perform check on the input  
+  // perform check on the input
  if (ux_input->type_num != NPY_FLOAT64){
    PyErr_Format(PyExc_TypeError, "`%s' only supports 64-bit float arrays for input array `%s`", Py_TYPE(self)->tp_name, forward_ux.name());
    return 0;
-  }  
+  }
  if (ux_input->ndim != 1){
    PyErr_Format(PyExc_TypeError, "`%s' only processes 1D arrays of float64 for `%s`", Py_TYPE(self)->tp_name, forward_ux.name());
    return 0;
-  }  
+  }
  if (ux_input->shape[0] != (Py_ssize_t)self->cxx->getNGaussians()*(Py_ssize_t)self->cxx->getNInputs()) {
    PyErr_Format(PyExc_TypeError, "`%s' 1D `input` array should have %" PY_FORMAT_SIZE_T "d, elements, not %" PY_FORMAT_SIZE_T "d for `%s`", Py_TYPE(self)->tp_name, (Py_ssize_t)self->cxx->getNGaussians()*(Py_ssize_t)self->cxx->getNInputs(), (Py_ssize_t)ux_input->shape[0], forward_ux.name());
    return 0;
  }
  double score = self->cxx->forward(*stats->cxx, *PyBlitzArrayCxx_AsBlitz<double,1>(ux_input));
  return Py_BuildValue("d", score);
  BOB_CATCH_MEMBER("cannot forward_ux", 0)
 }
@@ -564,7 +564,7 @@ static PyObject* PyBobLearnEMISVMachine_ForwardUx(PyBobLearnEMISVMachineObject*
 static auto forward = bob::extension::FunctionDoc(
  "forward",
  "Execute the machine",
-  "", 
+  "",
  true
 )
 .add_prototype("stats")
@@ -575,7 +575,7 @@ static PyObject* PyBobLearnEMISVMachine_Forward(PyBobLearnEMISVMachineObject* se
  char** kwlist = forward.kwlist(0);
  PyBobLearnEMGMMStatsObject* stats = 0;
  if (!PyArg_ParseTupleAndKeywords(args, kwargs, "O!", kwlist, &PyBobLearnEMGMMStats_Type, &stats))
    return 0;
@@ -607,14 +607,14 @@ static PyMethodDef PyBobLearnEMISVMachine_methods[] = {
    METH_VARARGS|METH_KEYWORDS,
    is_similar_to.doc()
  },
  {
    estimate_x.name(),
    (PyCFunction)PyBobLearnEMISVMachine_estimateX,
    METH_VARARGS|METH_KEYWORDS,
    estimate_x.doc()
  },
  {
    estimate_ux.name(),
    (PyCFunction)PyBobLearnEMISVMachine_estimateUx,

--- a/bob/learn/em/isv_trainer.cpp
+++ b/bob/learn/em/isv_trainer.cpp
@@ -88,7 +88,7 @@ int list_as_vector(PyObject* list, std::vector<blitz::Array<double,N> >& vec)
 static auto ISVTrainer_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".ISVTrainer",
  "ISVTrainer"
-  "References: [Vogt2008,McCool2013]",
+  "Train Intersession varibility modeling :ref:`ISV <isv>`.",
  ""
 ).add_constructor(
  bob::extension::FunctionDoc(

--- a/bob/learn/em/ivector_machine.cpp
+++ b/bob/learn/em/ivector_machine.cpp
@@ -15,8 +15,8 @@
 static auto IVectorMachine_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".IVectorMachine",
-  "An IVectorMachine consists of a Total Variability subspace :math:`T` and allows the extraction of IVector"
+  "Statistical model for the Total Variability training for more information and explanation see the user guide in documentation (:ref:`iVectors <ivector>`)" // this documentation text is intentionally written to be long!
-  "References: [Dehak2010]_",
+  "",
  ""
 ).add_constructor(
  bob::extension::FunctionDoc(
@@ -189,7 +189,7 @@ static auto supervector_length = bob::extension::VariableDoc(
  "Returns the supervector length.",
  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality"
-  "@warning An exception is thrown if no Universal Background Model has been set yet."
+  "An exception is thrown if no Universal Background Model has been set yet."
  ""
 );
 PyObject* PyBobLearnEMIVectorMachine_getSupervectorLength(PyBobLearnEMIVectorMachineObject* self, void*) {

--- a/bob/learn/em/ivector_trainer.cpp
+++ b/bob/learn/em/ivector_trainer.cpp
@@ -36,9 +36,9 @@ static int extract_GMMStats_1d(PyObject *list,
 static auto IVectorTrainer_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".IVectorTrainer",
  "IVectorTrainer"
-  "An IVectorTrainer to learn a Total Variability subspace :math:`$T$`"
+  "Trains the Total Variability subspace :math:`$T$` to generate :ref:`iVectors <ivector>`."
-  " (and eventually a covariance matrix :math:`$\\Sigma$`).",
+  "",
-  " References: [Dehak2010]"
+  ""
 ).add_constructor(
  bob::extension::FunctionDoc(
    "__init__",

--- a/bob/learn/em/jfa_base.cpp
+++ b/bob/learn/em/jfa_base.cpp
@@ -15,8 +15,8 @@
 static auto JFABase_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".JFABase",
-  "A JFABase instance can be seen as a container for :math:`U`, :math:`V` and :math:`D` when performing Joint Factor Analysis (JFA).\n\n"
+  "Container for :math:`U`, :math:`V` and :math:`D` when performing Joint Factor Analysis (:ref:`JFA <jfa>`).\n\n"
-  "References: [Vogt2008]_ [McCool2013]_",
+  "",
  ""
 ).add_constructor(
  bob::extension::FunctionDoc(
@@ -192,7 +192,7 @@ static auto supervector_length = bob::extension::VariableDoc(
  "Returns the supervector length.",
  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality"
-  "@warning An exception is thrown if no Universal Background Model has been set yet."
+  "An exception is thrown if no Universal Background Model has been set yet."
  ""
 );
 PyObject* PyBobLearnEMJFABase_getSupervectorLength(PyBobLearnEMJFABaseObject* self, void*) {

--- a/bob/learn/em/jfa_machine.cpp
+++ b/bob/learn/em/jfa_machine.cpp
@@ -173,8 +173,8 @@ static auto supervector_length = bob::extension::VariableDoc(
  "int",
  "Returns the supervector length.",
-  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality"
+  "NGaussians x NInputs: Number of Gaussian components by the feature dimensionality. "
-  "@warning An exception is thrown if no Universal Background Model has been set yet."
+  "An exception is thrown if no Universal Background Model has been set yet."
  ""
 );
 PyObject* PyBobLearnEMJFAMachine_getSupervectorLength(PyBobLearnEMJFAMachineObject* self, void*) {

--- a/bob/learn/em/jfa_trainer.cpp
+++ b/bob/learn/em/jfa_trainer.cpp
@@ -87,8 +87,8 @@ int list_as_vector(PyObject* list, std::vector<blitz::Array<double,N> >& vec)
 static auto JFATrainer_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".JFATrainer",
-  "JFATrainer"
+  "Trains a Joint Factor Analysis (:ref:`JFA <jfa>`) on top of GMMs"
-  "References: [Vogt2008,McCool2013]",
+  "",
  ""
 ).add_constructor(
  bob::extension::FunctionDoc(

--- a/bob/learn/em/kmeans_machine.cpp
+++ b/bob/learn/em/kmeans_machine.cpp
@@ -15,7 +15,7 @@
 static auto KMeansMachine_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".KMeansMachine",
-  "This class implements a k-means classifier.\n"
+  "Statistical model for the :ref:`k-means <kmeans>` .\n"
  "See Section 9.1 of Bishop, \"Pattern recognition and machine learning\", 2006"
 ).add_constructor(
  bob::extension::FunctionDoc(

--- a/bob/learn/em/kmeans_trainer.cpp
+++ b/bob/learn/em/kmeans_trainer.cpp
@@ -42,8 +42,8 @@ static inline const std::string& IM2string(bob::learn::em::KMeansTrainer::Initia
 static auto KMeansTrainer_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".KMeansTrainer",
-  "Trains a KMeans machine."
+  "Trains a KMeans clustering :ref:`k-means <kmeans>`."
-  "This class implements the expectation-maximization algorithm for a k-means machine."
+  "This class implements the expectation-maximization algorithm for a k-means."
  "See Section 9.1 of Bishop, \"Pattern recognition and machine learning\", 2006"
  "It uses a random initialization of the means followed by the expectation-maximization algorithm"
 ).add_constructor(

--- a/bob/learn/em/linear_scoring.cpp
+++ b/bob/learn/em/linear_scoring.cpp
@@ -71,6 +71,7 @@ static inline bool f(PyObject* o){return o != 0 && PyObject_IsTrue(o) > 0;}
 /*** linear_scoring ***/
 bob::extension::FunctionDoc linear_scoring1 = bob::extension::FunctionDoc(
  "linear_scoring",
+  "The :ref:`Linear scoring <linearscoring>` is an approximation to the log-likelihood ratio that was shown to be as accurate and up to two orders of magnitude more efficient to compute [Glembek2009]_."
  "",
  0,
  true

--- a/bob/learn/em/map_gmm_trainer.cpp
+++ b/bob/learn/em/map_gmm_trainer.cpp
@@ -17,7 +17,7 @@ static inline bool f(PyObject* o){return o != 0 && PyObject_IsTrue(o) > 0;}  /*
 static auto MAP_GMMTrainer_doc = bob::extension::ClassDoc(
  BOB_EXT_MODULE_PREFIX ".MAP_GMMTrainer",
-  "This class implements the maximum a posteriori M-step of the expectation-maximization algorithm for a GMM Machine. The prior parameters are encoded in the form of a GMM (e.g. a universal background model). The EM algorithm thus performs GMM adaptation."
+  "This class implements the maximum a posteriori (:ref:`MAP <map>`) M-step of the expectation-maximization algorithm for a GMM Machine. The prior parameters are encoded in the form of a GMM (e.g. a universal background model). The EM algorithm thus performs GMM adaptation."
 ).add_constructor(
  bob::extension::FunctionDoc(
    "__init__",
No results found