On Fri, Nov 27, 2009 at 5:59 PM, Dag Sverre Seljebotn <da...@student.matnat.uio.no> wrote: > William Stein wrote: >> On Fri, Nov 27, 2009 at 7:37 AM, Dag Sverre Seljebotn >> <da...@student.matnat.uio.no> wrote: >>> I need a convenient matrix class for some numerical work. SciPy doesn't >>> accept GPL code and besides I heavily dislike the NumPy matrix class. So >>> why not put my efforts into improving Sage (or at least write something >>> Sage-compatible for myself)... >> >> Cool. Make sure to also look into CVXOPT >> (http://abel.ee.ucla.edu/cvxopt/) which is included in Sage and does >> accept GPL code. > > Yep, that's what I use now. > >>> 2) I haven't checked but I think converting all my sparse values to a >>> Python dict to get it into a Sage matrix would be too slow, at least it >>> is wasteful. So I'd like something like >>> >>> matrix(CDF, I=I, J=J, V=V, sparse=True) >>> >>> where I, J, and V are vectors containing respectively rows, columns, and >>> values, element by element. >> >> I don't quite understand this yet. Note that you can directly access >> your Matrix_sparse_rdf from Cython, if you so choose. > > Yes, but it would be nice to have an efficient way of constructing a > sparse matrix from Python-space as well. Both SciPy and CVXOPT supports > passing three vectors (with row indices, col indices, and values) to > initialize a sparse matrix. For instance, the 3x3 identity matrix would be > > matrix(RDF, I=[0, 1, 2], J=[0, 1, 2], V=[1, 1, 1], sparse=True) > > (Where we could use different keywords for I, J, V, or combine them in a > tuple). > > The point is that I, J and V could be NumPy arrays or vector instances, > which would allow for efficient construction without creating millions > of Python objects in a dictionary. > > The indices need not be sorted, and in fact in NumPy you can pass the > same index pair many times (which results in summation over > corresponding values, which they claim is useful sometimes). >
Well I think it would be great for you to implement something like that. William -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
