[Scipy-svn] r3075 - in trunk/Lib/cluster: . tests

[scipy-svn at scipy.org]

Author: cdavid
Date: 2007-06-07 23:09:24 -0500 (Thu, 07 Jun 2007)
New Revision: 3075

Modified:
   trunk/Lib/cluster/tests/test_vq.py
   trunk/Lib/cluster/vq.py
Log:
Modify kmeans2 arguments order so that they conform to the ones from kmeans

Modified: trunk/Lib/cluster/tests/test_vq.py
===================================================================
--- trunk/Lib/cluster/tests/test_vq.py	2007-06-08 02:36:11 UTC (rev 3074)
+++ trunk/Lib/cluster/tests/test_vq.py	2007-06-08 04:09:24 UTC (rev 3075)
@@ -1,7 +1,7 @@
 #! /usr/bin/env python
 
 # David Cournapeau
-# Last Change: Sat May 05 06:00 PM 2007 J
+# Last Change: Fri Jun 08 12:00 PM 2007 J
 
 # For now, just copy the tests from sandbox.pyem, so we can check that
 # kmeans works OK for trivial examples.
@@ -72,6 +72,7 @@
     #    print _py_vq_1d(data, initc)
 
 class test_kmean(NumpyTestCase):
+    #def check_kmeans
     def check_kmeans_simple(self, level=1):
         initc = N.concatenate(([[X[0]], [X[1]], [X[2]]]))
         code = initc.copy()
@@ -93,8 +94,8 @@
         """Testing simple call to kmeans2 and its results."""
         initc = N.concatenate(([[X[0]], [X[1]], [X[2]]]))
         code = initc.copy()
-        code1 = kmeans2(X, code, niter = 1)[0]
-        code2 = kmeans2(X, code, niter = 2)[0]
+        code1 = kmeans2(X, code, iter = 1)[0]
+        code2 = kmeans2(X, code, iter = 2)[0]
 
         assert_array_almost_equal(code1, CODET1)
         assert_array_almost_equal(code2, CODET2)

Modified: trunk/Lib/cluster/vq.py
===================================================================
--- trunk/Lib/cluster/vq.py	2007-06-08 02:36:11 UTC (rev 3074)
+++ trunk/Lib/cluster/vq.py	2007-06-08 04:09:24 UTC (rev 3075)
@@ -466,7 +466,7 @@
 
 _valid_init_meth = {'random': _krandinit, 'points': _kpoints}
 
-def kmeans2(data, k, minit='random', niter=10):
+def kmeans2(data, k, iter = 10, thresh = 1e-5, minit='random'):
     """Classify a set of points into k clusters using kmean algorithm.
 
     The algorithm works by minimizing the euclidian distance between data points
@@ -481,6 +481,10 @@
         k : int or ndarray
             Number of clusters. If a ndarray is given instead, it is
             interpreted as initial cluster to use instead.
+        niter : int
+            Number of iterations to run.
+        niter : float
+            (not used yet).
         minit : string
             Method for initialization. Available methods are random, points and
             uniform:
@@ -493,9 +497,6 @@
             uniform choses k points from the data such are they form a uniform
             grid od the dataset.
 
-        niter : int
-            Number of iterations to run.
-
     :Returns:
         clusters : ndarray
             the found clusters (one cluster per row).
@@ -535,8 +536,8 @@
             raise ValueError("unknown init method %s" % str(minit))
         clusters = init(data, k)
 
-    assert not niter == 0
-    return _kmeans2(data, clusters, niter, nc)
+    assert not iter == 0
+    return _kmeans2(data, clusters, iter, nc)
 
 def _kmeans2(data, code, niter, nc):
     """ "raw" version of kmeans2. Do not use directly.