[MRG] FIX Fixes failure in interpolation of ADASYN (scikit-learn-cont…

…rib#235) * Fixed ADASYN * Added bug fix to list * Updated tests
glemaitre · Feb 21, 2017 · 722e223 · 722e223
1 parent 5413a29
commit 722e223
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 5 deletions.
diff --git a/doc/whats_new.rst b/doc/whats_new.rst
@@ -25,6 +25,7 @@ Bug fixes
 - Fixed a bug in :class:`pipeline.Pipeline`, solve the issue to put to sampler in the same `Pipeline`. By `Christos Aridas`_ .
 - Fixed a bug in :class:`under_sampling.CondensedNeareastNeigbour`, correction of the shape of `sel_x` when only one sample is selected. By `Aliaksei Halachkin`_.
 - Fixed a bug in :class:`under_sampling.NeighbourhoodCleaningRule`, selecting neighbours instead of minority class misclassified samples. By `Aleksandr Loskutov`_.
+- Fixed a bug in :class:`over_sampling.ADASYN`, correction of the creation of a new sample so that the new sample lies between the minority sample and the nearest neighbour. By `Rafael Wampfler`_.
 
 New features
 ~~~~~~~~~~~~
@@ -119,3 +120,4 @@ New methods
 .. _Francois Magimel: https://github.com/Linkid
 .. _Aliaksei Halachkin: https://github.com/honeyext
 .. _Aleksandr Loskutov: https://github.com/loskutyan
+.. _Rafael Wampfler: https://github.com/Eichhof
diff --git a/imblearn/over_sampling/adasyn.py b/imblearn/over_sampling/adasyn.py
@@ -217,7 +217,7 @@ def _sample(self, X, y):
             # Create a new sample
             for nn_z in nn_zs:
                 step = random_state.uniform()
-                x_gen = x_i + step * (x_i - X[x_i_nn[nn_z], :])
+                x_gen = x_i + step * (X[x_i_nn[nn_z], :] - x_i)
                 X_resampled = np.vstack((X_resampled, x_gen))
                 y_resampled = np.hstack((y_resampled, self.min_c_))
 

diff --git a/imblearn/over_sampling/tests/test_adasyn.py b/imblearn/over_sampling/tests/test_adasyn.py
@@ -127,8 +127,8 @@ def test_ada_fit_sample():
                      [-0.18410027, -0.45194484], [0.9281014, 0.53085498],
                      [-0.14374509, 0.27370049], [-0.41635887, -0.38299653],
                      [0.08711622, 0.93259929], [1.70580611, -0.11219234],
-                     [0.29427267, 0.21740707], [0.68118697, -0.25220353],
-                     [1.37180201, 0.37279378], [-0.59243851, -0.80715327]])
+                     [-0.06182085, -0.28084828], [0.38614986, -0.35405599],
+                     [0.39635544, 0.33629036], [-0.24027923, 0.04116021]])
     y_gt = np.array([
         0, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 0, 0, 0
     ])
@@ -204,8 +204,8 @@ def test_ada_fit_sample_nn_obj():
                      [-0.18410027, -0.45194484], [0.9281014, 0.53085498],
                      [-0.14374509, 0.27370049], [-0.41635887, -0.38299653],
                      [0.08711622, 0.93259929], [1.70580611, -0.11219234],
-                     [0.29427267, 0.21740707], [0.68118697, -0.25220353],
-                     [1.37180201, 0.37279378], [-0.59243851, -0.80715327]])
+                     [-0.06182085, -0.28084828], [0.38614986, -0.35405599],
+                     [0.39635544, 0.33629036], [-0.24027923, 0.04116021]])
     y_gt = np.array([
         0, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 0, 0, 0
     ])