Documentation updated to version 1.7.2

CHANGES

@@ -1,4 +1,4 @@
-Since 1.7.1:
+Version 1.7.2: 2016-11-24
   - New: The coefficient of the Gaussian neighborhood function exp(-||x-y||^2/(2*(coeff*radius)^2)) is now exposed in all interfaces as a parameter.
   - New: `get_bmu` function in the Python interface to get the best matching units given an activation map.
   - Changed: Updated PCA initialization in the Python interface to work with `sk-learn` 0.18 onwards.

src/Python/doc/source/example.rst

@@ -24,7 +24,7 @@ First, we import the necessary modules:
     import matplotlib.pyplot as plt
     from mpl_toolkits.mplot3d import Axes3D
     import somoclu
-    %matplotlib inline  
+    %matplotlib inline
 
 Then we generate and plot some random data in three categories:
 
@@ -56,7 +56,7 @@ that qualifies as an emergent self-organizing map for this data:
 .. code:: python
 
     n_rows, n_columns = 100, 160
-    som = somoclu.Somoclu(n_columns, n_rows, data=data)
+    som = somoclu.Somoclu(n_columns, n_rows, data=data, compactsupport=False)
     %time som.train()
 
 
@@ -119,7 +119,7 @@ lower right corner:
 
 .. code:: python
 
-    som.view_umatrix(bestmatches=True, bestmatchcolors=colors, labels=labels, 
+    som.view_umatrix(bestmatches=True, bestmatchcolors=colors, labels=labels,
                      zoom=((50, n_rows), (100, n_columns)))
 
 
@@ -143,7 +143,8 @@ type as follows:
 
 .. code:: python
 
-    som = somoclu.Somoclu(n_columns, n_rows, data=data, maptype="toroid")
+    som = somoclu.Somoclu(n_columns, n_rows, data=data, maptype="toroid",
+                          compactsupport=False)
     som.train()
     som.view_umatrix(bestmatches=True, bestmatchcolors=colors)
 
@@ -165,7 +166,8 @@ neurons are also implemented:
 
 .. code:: python
 
-    som = somoclu.Somoclu(n_columns, n_rows, data=data, gridtype="hexagonal")
+    som = somoclu.Somoclu(n_columns, n_rows, data=data, gridtype="hexagonal",
+                          compactsupport=False)
     som.train()
     som.view_umatrix(bestmatches=True, bestmatchcolors=colors)
 
@@ -195,7 +197,8 @@ this, we need to pass an optional argument to the constructor:
 
 .. code:: python
 
-    som = somoclu.Somoclu(n_columns, n_rows, data=data, maptype="toroid", initialization="pca")
+    som = somoclu.Somoclu(n_columns, n_rows, data=data, maptype="toroid",
+                          compactsupport=False, initialization="pca")
     som.train()
     som.view_umatrix(bestmatches=True, bestmatchcolors=colors)
 

src/Python/doc/source/reference.rst

@@ -2,7 +2,7 @@
 Function Reference
 ******************
 
-.. py:class:: Somoclu(n_columns, n_rows, data=None, initialcodebook=None, kerneltype=0, maptype='planar', gridtype='rectangular', compactsupport=False, neighborhood='gaussian', initialization=None)
+.. py:class:: Somoclu(n_columns, n_rows, data=None, initialcodebook=None, kerneltype=0, maptype='planar', gridtype='rectangular', compactsupport=False, neighborhood='gaussian', std_coeff=0.5, initialization=None)
 
    Class for training and visualizing a self-organizing map.
 
@@ -77,6 +77,19 @@ Function Reference
       :returns: The the dot product of the codebook and the data.
       :rtype: 2D numpy.array
 
+
+   .. py:method:: Somoclu.get_bmus(activation_map)
+
+      Return Best Matching Unit indexes of the activation map.
+
+      :param activation_map: Activation map computed with self.get_surface_state()
+      :type activation_map: 2D numpy.array
+
+      :returns: The bmus indexes corresponding to this activation map
+                (same as self.bmus for the training samples).
+      :rtype: 2D numpy.array
+
+
    .. py:method:: Somoclu.load_bmus(filename)
 
       Load the best matching units from a file to the Somoclu object.
@@ -118,9 +131,9 @@ Function Reference
                                  * "linear": Linear interpolation (default)
                                  * "exponential": Exponential decay
       :param scale0: The initial learning scale. Default value: 0.1.
-      :type scale0: int.
+      :type scale0: float.
       :param scaleN: The learning scale in the final epoch. Default: 0.01.
-      :type scaleN: int.
+      :type scaleN: float.
       :param scalecooling: The cooling strategy between scale0 and scaleN:
 
                                  * "linear": Linear interpolation (default)