Hello,
I've just spent an enjoyable evening fighting with CVXOPT and SAGE.
The results are 2 of the examples from Convex Optimization by Boyd and
Vandenberghe.
http://www.stanford.edu/~boyd/cvxbook/ almost
implemented. It is messy, but published at
http://alpha.sagenb.org/home/pub/17/.
I post it here incase it is of any interest, and to see if there are
any suggestions on, a) improving the interface between SAGE matrices
and CVXOPT, b) how to update the CVXOPT package as it now seems to be
out of date, and does not match the CVXOPT documentation?
Regards
Martin.