update DBSCAN

src/impetuous/clustering.py

@@ -550,11 +550,11 @@ def backprojection_clustering ( analyte_df , bRanked=False , n_dimensions=2 ,
     return ( cluster_coords_f,cluster_coords_s )
 
 def dbscan ( data_frame = None , distance_matrix = None ,
-        eps = 0.11, minPts = 1 ) :
+        eps = None, minPts = None ) :
     #
     # FOR A DESCRIPTION OF THE CONNECTIVITY READ PAGE 30 (16 INTERNAL NUMBERING) of:
     # https://kth.diva-portal.org/smash/get/diva2:748464/FULLTEXT01.pdf
-    from impetuous.clustering import connectivity , absolute_coordinates_to_distance_matrix
+    from impetuous.clustering import absolute_coordinates_to_distance_matrix
     import operator
     if not operator.xor( data_frame is None , distance_matrix is None ) :
         print ( "ONLY SUPPLY A SINGE DATA FRAME OR A DISTANCE MATRIX" )
@@ -564,17 +564,28 @@ def dbscan ( data_frame = None , distance_matrix = None ,
     if not data_frame is None :
         if not 'pandas' in str(type(data_frame)) :
             print ( "ONLY SUPPLY A SINGE DATA FRAME WITH ABSOLUTE COORDINATES" )
-            print ( "DATA MATRICES NEEDS TO BE SPECIFIED WITH \" distance_matrix = ... \" " )
+            print ( "DATA MATRICES NEEDS TO BE SPECIFIED WITH \" data_frame = ... \" " )
             print ( "dbscan FAILED" )
             exit ( 1 )
         if bVerbose :
             print ( data_frame )
+    if not ( 'float' in str(type(eps)).lower() and 'int' in str(type(minPts)).lower() ) :
+        print ( "TO CALL THE dbscan PLEASE SPECIFY AT LEAST A DATA FRAME OR")
+        print ( "ITS CORRESPONDING DISTANCE MATRIX AS WELL AS THE DISTANCE CUTOFF PARAMETER" )
+        print ( "AND THE MINIMAL AMOUNT OF NEIGHBOUR POINTS TO CONSIDER IT A CLUSTER")
+        print ( "dbscan ( data_frame = None , distance_matrix = None , eps = None, minPts = None )" )
         distance_matrix = absolute_coordinates_to_distance_matrix(data_frame.values)
-    clustercontent , clustercontacts  =  connectivity(distance_matrix,eps)
     isNoise = np.sum(distance_matrix<eps,0)-1 < minPts
+    i_ = 0
+    for ib in isNoise :
+        if ib :
+            distance_matrix [ i_] = ( 1+eps )*10.0
+            distance_matrix.T[i_] = ( 1+eps )*10.0
+            distance_matrix[i_][i_] = 0.
+        i_ = i_+1
+    clustercontent , clustercontacts  =  connectivity(distance_matrix,eps)
     return ( {'cluster content': clustercontent, 'clusterid-particleid' : clustercontacts, 'is noise':isNoise} )
 
-
 if __name__ == '__main__' :
 
     if False :