GraphLab: Diagonally dominant matrix

14 views
Skip to first unread message

Verónica Vergara Larrea (SCC13 Science Lead)

unread,
Nov 19, 2013, 4:19:55 PM11/19/13
to student-cluste...@googlegroups.com
A team asked for more info for Task 2 of GraphLab in which the matrix must be scaled. The short answer, is yes, you need to modify the input matrix in your code.

The Application Specialist confirms that it is ok to change the diagonal to make the input diagonally dominant. As the instructions mention this can be done by setting the value of the diagonal to the sum of the respective row+1. 

Best,
Verónica

Verónica Vergara Larrea (SCC13 Science Lead)

unread,
Nov 19, 2013, 7:25:37 PM11/19/13
to student-cluste...@googlegroups.com
For another example of what "scaling the matrix" means, here is the sample for the before and after scaling for the 3x3 matrix:

A=[ 1.8147 0.9134 0.2785 
      0.9058 1.6324 0.5469 
      0.1270 0.0975 1.9575 ]; 
b= [ 1.0 1.0 1.0 ]'; 

The result after making the matrix diagonaly dominant (+1) should be: 
>> A2 = (A­diag(diag(A)))+ diag(sum(A'))+eye(3) 
A2 = 4.0066 0.9134 0.2785 
        0.9058 4.0851 0.5469 
        0.1270 0.0975 3.1820
Reply all
Reply to author
Forward
0 new messages