Improved tests for PCA and LDA

bob/bio/base/algorithm/Algorithm.py

@@ -49,8 +49,8 @@ class Algorithm:
     self.split_training_features_by_client = split_training_features_by_client
     self.use_projected_features_for_enrollment = performs_projection and use_projected_features_for_enrollment
     self.requires_enroller_training = requires_enroller_training
-    self.m_model_fusion_function = utils.score_fusion_strategy(multiple_model_scoring)
-    self.m_probe_fusion_function = utils.score_fusion_strategy(multiple_probe_scoring)
+    self.model_fusion_function = utils.score_fusion_strategy(multiple_model_scoring)
+    self.probe_fusion_function = utils.score_fusion_strategy(multiple_probe_scoring)
     self._kwargs = kwargs
     self._kwargs.update({'multiple_model_scoring':multiple_model_scoring, 'multiple_probe_scoring':multiple_probe_scoring})
 
@@ -80,9 +80,9 @@ class Algorithm:
     and fuses the scores using the fusion method specified in the constructor of this class.
     Usually this function is called from derived class 'score' functions."""
     if isinstance(models, list):
-      return self.m_model_fusion_function([self.score(model, probe) for model in models])
+      return self.model_fusion_function([self.score(model, probe) for model in models])
     elif isinstance(models, numpy.ndarray):
-      return self.m_model_fusion_function([self.score(models[i,:], probe) for i in range(models.shape[0])])
+      return self.model_fusion_function([self.score(models[i,:], probe) for i in range(models.shape[0])])
     else:
       raise ValueError("The model does not have the desired format (list, array, ...)")
 
@@ -92,7 +92,7 @@ class Algorithm:
     In this base class implementation, it computes the scores for each probe file using the 'score' method,
     and fuses the scores using the fusion method specified in the constructor of this class."""
     if isinstance(probes, list):
-      return self.m_probe_fusion_function([self.score(model, probe) for probe in probes])
+      return self.probe_fusion_function([self.score(model, probe) for probe in probes])
     else:
       # only one probe feature -> use the default scoring function
       return self.score(model, probes)

bob/bio/base/algorithm/LDA.py

@@ -54,10 +54,13 @@ class LDA (Algorithm):
     self.uses_variances = uses_variances
 
 
-  def _check_feature(self, feature):
+  def _check_feature(self, feature, projected=False):
     """Checks that the features are appropriate"""
     if not isinstance(feature, numpy.ndarray) or len(feature.shape) != 1 or feature.dtype != numpy.float64:
       raise ValueError("The given feature is not appropriate")
+    index = 1 if projected else 0
+    if self.machine is not None and feature.shape[0] != self.machine.shape[index]:
+      raise ValueError("The given feature is expected to have %d elements, but it has %d" % (self.machine.shape[index], feature.shape[0]))
 
 
   def _arrange_data(self, training_files):
@@ -164,13 +167,14 @@ class LDA (Algorithm):
   def enroll(self, enroll_features):
     """Enrolls the model by storing all given input vectors"""
     assert len(enroll_features)
-    [self._check_feature(feature) for feature in enroll_features]
+    [self._check_feature(feature, True) for feature in enroll_features]
     # just store all the features
     return numpy.vstack(enroll_features)
 
 
   def score(self, model, probe):
     """Computes the distance of the model to the probe using the distance function"""
+    self._check_feature(probe, True)
     # return the negative distance (as a similarity measure)
     if len(model.shape) == 2:
       # we have multiple models, so we use the multiple model scoring

bob/bio/base/algorithm/PCA.py

@@ -46,10 +46,13 @@ class PCA (Algorithm):
     self.uses_variances = uses_variances
 
 
-  def _check_feature(self, feature):
+  def _check_feature(self, feature, projected=False):
     """Checks that the features are appropriate"""
     if not isinstance(feature, numpy.ndarray) or len(feature.shape) != 1 or feature.dtype != numpy.float64:
       raise ValueError("The given feature is not appropriate")
+    index = 1 if projected else 0
+    if self.machine is not None and feature.shape[0] != self.machine.shape[index]:
+      raise ValueError("The given feature is expected to have %d elements, but it has %d" % (self.machine.shape[index], feature.shape[0]))
 
 
   def train_projector(self, training_features, projector_file):
@@ -104,13 +107,14 @@ class PCA (Algorithm):
   def enroll(self, enroll_features):
     """Enrolls the model by storing all given input vectors"""
     assert len(enroll_features)
-    [self._check_feature(feature) for feature in enroll_features]
+    [self._check_feature(feature, True) for feature in enroll_features]
     # just store all the features
     return numpy.vstack(enroll_features)
 
 
   def score(self, model, probe):
     """Computes the distance of the model to the probe using the distance function"""
+    self._check_feature(probe, True)
     # return the negative distance (as a similarity measure)
     if len(model.shape) == 2:
       # we have multiple models, so we use the multiple model scoring