diff --git a/bob/pad/base/algorithm/PadLDA.py b/bob/pad/base/algorithm/PadLDA.py
index 91dc05d01f64d978e586804772cc030c704bc04b..d84efc455c82f951083d3dcbc38da0404cb7c088 100644
--- a/bob/pad/base/algorithm/PadLDA.py
+++ b/bob/pad/base/algorithm/PadLDA.py
@@ -8,8 +8,8 @@ class PadLDA(LDA):
     """Wrapper for bob.bio.base.algorithm.LDA,
     
     Here, LDA is used in a PAD context. This means that the feature
-    will be projected on a two-dimensional subspace, where the two
-    dimensions represents the real and attack classes.
+    will be projected on a single dimension subspace, which acts as a score
+
 
     For more details, you may want to have a look at 
     `bob.learn.linear Documentation`_
diff --git a/bob/pad/base/test/test_algorithms.py b/bob/pad/base/test/test_algorithms.py
index f73b0f06789b94205070e7bce7fda49993bd09ce..1e4d9106aa67f6f99e0734cc3dd0d97d80f987a8 100644
--- a/bob/pad/base/test/test_algorithms.py
+++ b/bob/pad/base/test/test_algorithms.py
@@ -13,6 +13,7 @@ import bob.pad.base
 from bob.pad.base.algorithm import SVM
 from bob.pad.base.algorithm import OneClassGMM
 from bob.pad.base.algorithm import MLP
+from bob.pad.base.algorithm import PadLDA
 
 import random
 
@@ -231,8 +232,4 @@ def test_LDA():
 
     lda = PadLDA()
     lda.train_projector(training_features, '/tmp/lda.hdf5')
-
-    real_sample = real_array[0]
-    prob = lda.project(real_sample)
-    assert prob[0] > prob[1]
-
+    assert lda.machine.shape == (2, 1)