Geek Answers Handbook

F # 1.9.9.9 and Lapak

Can someone help me with an example using lapack with the latest f #? Or not recommended yet?

I searched and found http://fdatamining.blogspot.com/ , but could not compile FSharp.PowerPack.Math.Providers.dll .

EDIT: @Yin, Thanks for your blog. When compiling FSharp.PowerPack.Math.Providers.dll from source 1.9.7.8, Microsoft.FSharp.Compatibility.permutation used several times and cannot be found. Your blog says just define it explicitly:

 type permutation = int –> int.

Alternatively, I replace all Microsoft.FSharp.Compatibility.permutation with Microsoft.FSharp.Math.Permutation and compiled it. Not tested yet.

Could you show us a test case for svd error?

thanks

+6
math f # lapack
source share
1 answer

I wrote this blog. Are you using VS 2008? Have you installed PowerPack? What have you done so far? What compilation errors did you get?

Using math is not recommended with the F # command. However, this is the best I can find on the Internet for F # (the Math.Net incomplete and alpha test takes time to mature). I can also tell you that a buggy wrapper, for example. There is an error in the SVD wrapper. But this shell is small, you can maintain it yourself on new F # releases and add new features as needed (we have an obvious license problem here!). My plan is to use and modify this shell until it becomes stable, and rewrite the new one to fix the license problem. Or we can convince MS or MSR to continue development.

@ahala, svd error

Search member this.dgesvd _ ((a: matrix)) = in lapack_service_netlib.fs

This function has a memory / segment access error when performing SVD on the m-by-n matrix (m> n).

To solve the problem, we need to change arg_jobu and arg_jobvt to 'S'. ref lapack doc here: http://linux.die.net/man/l/dgesvd

+3
source share

More articles:


All Articles