For almost anything math related, PDL is the right choice in perl. Where I work, we've used PDL for harmonic analysis and linear regression on large data sets. We recently added some graphing capabilities that utilize PDL+GD . We are also exploring adding neural net predictive capabilities to our toolset using PDL. Hard core math-types balk at perl as being "slow", but PDL is anything but. It adds much of the same functionality as you would find in matlab to perl (The big advantage that matlab has over PDL are those toolboxes that you can get for matlab. PDL has nothing analogous yet, but given CPAN, it's only a matter of time). Where perl really wins is getting the data into and out of whatever format you need for your calculations. Matlab, though sophisticated, is still almost as clumsy as Fortran in this regard.
|