Merge pull request #136 from H-Dempsey/master

Fixing an issue with agreement_weighted

Author: aestrivex

bct/algorithms/clustering.py

@@ -73,7 +73,7 @@ def agreement_weighted(ci, wts):
 
     Parameters
     ----------
-    ci : MxN np.ndarray
+    ci : NxM np.ndarray
         set of M (possibly degenerate) partitions of N nodes
     wts : Mx1 np.ndarray
         relative weight of each partition
@@ -84,12 +84,12 @@ def agreement_weighted(ci, wts):
         weighted agreement matrix
     '''
     ci = np.array(ci)
-    m, n = ci.shape
+    n, m = ci.shape
     wts = np.array(wts) / np.sum(wts)
 
     D = np.zeros((n, n))
     for i in range(m):
-        d = dummyvar(ci[i, :].reshape(1, n))
+        d = dummyvar(ci[:, i].reshape(n, 1))
         D += np.dot(d, d.T) * wts[i]
     return D
 

test/clustering_test.py

@@ -105,8 +105,26 @@ def test_transitivity_bd():
 
 
 def test_agreement_weighted():
-    # this function is very hard to use or interpret results from
-    pass
+    # Test whether agreement gives the same results as 
+    # agreement_weighted when the weights are all the same
+    ci = np.array([[1, 1, 2, 2, 3],
+                   [1, 2, 2, 3, 3],
+                   [1, 1, 2, 3, 3]]).T
+    wts = np.ones(ci.shape[1])
+    
+    D_agreement = bct.agreement(ci)
+    D_weighted = bct.agreement_weighted(ci, wts)
+    
+    # Undo the normalization and fill the diagonal with zeros 
+    # in D_weighted to get the same result as D_agreement
+    D_weighted = D_weighted * ci.shape[1]
+    np.fill_diagonal(D_weighted, 0)
+
+    print('agreement matrix:')
+    print(D_agreement)
+    print('weighted agreement matrix:')
+    print(D_weighted)
+    assert (D_agreement == D_weighted).all()
 
 def test_agreement():
     # Case 1: nodes > partitions