In the next couple of weeks there will be a nuget package that includes the MKL native provider. In the mean time, here are links to them:
http://files.marcuscuda.com/mathnet.native.x86-64.zip - 64bit
http://files.marcuscuda.com/mathnet.native.x86.zip - 32bit
Put the two DLLs in the same directory as your executable and then add
Control.LinearAlgebraProvider = new MathNet.Numerics.Algorithms.LinearAlgebra.Mkl.MklLinearAlgebraProvider();
to your code.
Let me know how it performs.
Hi Marcus.
I downloaded those 2 DLL's to my main project folder, then I added that line you mentioned. Although, when I started debugging, Visual Studio poped up a window to browse a .CS file to add to the project. I guess i should just import the DLLs somehow, tried also to add them as references, but VS won't let me do that, saying that they're not "a valid assembly or COM component"... Any thoughts ?
Hi,
You don't add them as a reference. You need to copy them to the output folder (i.e.:<project>\bin\release). What I normally do is add them to the project (Add -> Existing item) and then set the "Copy to output directory" property to "Copy Always" for both files.
WOW !
It just flies! As I told you, on my mac it was taking about 2 or 3 seconds per iteration, while with mathnetnumerics it took about 1 min. Now it is performing like Fortran's LAPACK or even better. It's actually hard to tell once it's so fast ! :)
Problem solved... THANK YOU ! :)
I have a trouble. When i use a big sparse matrix for QR factorization it fails and rises OutOfRange Exception. Could you help me please?
SparseMatrix m = new SparseMatrix(....); rows count = column count and grater than 1000
var qr = m.QR(); --- Fails
System.ArgumentOutOfRangeException was unhandled by user code
Message=Non-negative number required.
Parameter name: length
Source=mscorlib
ParamName=length
StackTrace:
at System.Array.BinarySearch[T](T[] array, Int32 index, Int32 length, T value, IComparer`1 comparer)
at System.Array.BinarySearch[T](T[] array, Int32 index, Int32 length, T value)
at MathNet.Numerics.LinearAlgebra.Storage.SparseCompressedRowMatrixStorage`1.FindItem(Int32 row, Int32 column) in c:\TeamCity\buildAgent\work\d4ecde2945c804d6\src\Numerics\LinearAlgebra\Storage\SparseCompressedRowMatrixStorage.cs:line
180
at MathNet.Numerics.LinearAlgebra.Storage.SparseCompressedRowMatrixStorage`1.At(Int32 row, Int32 column, T value) in c:\TeamCity\buildAgent\work\d4ecde2945c804d6\src\Numerics\LinearAlgebra\Storage\SparseCompressedRowMatrixStorage.cs:line
77
at MathNet.Numerics.LinearAlgebra.Generic.Matrix`1.At(Int32 row, Int32 column, T value) in c:\TeamCity\buildAgent\work\d4ecde2945c804d6\src\Numerics\LinearAlgebra\Generic\Matrix.cs:line 204
at MathNet.Numerics.LinearAlgebra.Double.Factorization.UserQR.ComputeQR(Double[] u, Matrix`1 a, Int32 rowStart, Int32 rowDim, Int32 columnStart, Int32 columnDim, Int32 availableCores) in c:\TeamCity\buildAgent\work\d4ecde2945c804d6\src\Numerics\LinearAlgebra\Double\Factorization\UserQR.cs:line
218
at MathNet.Numerics.LinearAlgebra.Double.Factorization.UserQR.<>c__DisplayClass6.<ComputeQR>b__3() in c:\TeamCity\buildAgent\work\d4ecde2945c804d6\src\Numerics\LinearAlgebra\Double\Factorization\UserQR.cs:line
204
at System.Threading.Tasks.Task.InnerInvoke()
at System.Threading.Tasks.Task.Execute()
InnerException:
Hi,
will there be support for IR filter in Math.NET Numerics?
I know there is Math.NET Neodym, but it depends on the discontinued Math.NET Iridium. It would be great to have the filters on top of Math.NET Numerics.
Yes, it would make sense to port them over (also all the windowing functions); we already have ported some sample generators to the .Signals namespace.
What specific filters would you need? Also filter coefficient designers/calculators or are they already pre-computed in your case?
Thanks,
Christoph
PS: funny how people continue to cross-post to Meta Numerics ;) (no problem, just observed it more often recently)
We are interested in IIR filters for which we have pre-computed coefficients (no need for coefficient calculators ala butter()). The zero-phase filter filtfilt() would be really helpful, too.
Thanks,
Thomas
I'm not a math expert and need a Simplex optimization function, is there an example about this?
Thanks
Alessandro
Hi,
it may be a dump question but I've a program which is written in C and uses the dgesvd function of netlib (http://www.netlib.org/lapack/double/dgesvd.f).
Now I want to rewrite what is done in the C program in C# and tried to use Math.Net Numerics Double.DenseSvd and I don't get the same results.
Can someone tell me what is the difference between both implementations?
I used the following input matrix:
DenseMatrix A = new DenseMatrix(24, 3);
A[0, 0] = 0.00853749999998854; A[0, 1] = 89.2522666666667; A[0, 2] = 0;
A[1, 0] = -0.0175625000000537; A[1, 1] = 55.1682666666667; A[1, 2] = 0;
A[2, 0] = -0.0174625000000788; A[2, 1] = 50.8299666666667; A[2, 2] = 0;
A[3, 0] = 0.000137499999937063; A[3, 1] = 32.6130666666667; A[3, 2] = 0;
A[4, 0] = 0.025837499999966; A[4, 1] = 2.34146666666666; A[4, 2] = 0;
A[5, 0] = 0.0247375000000147; A[5, 1] = 0.0117666666666638; A[5, 2] = 0;
A[6, 0] = 0.00273749999996653; A[6, 1] = -17.3512333333333; A[6, 2] = 0;
A[7, 0] = -0.0143625000000611; A[7, 1] = -27.1400333333333; A[7, 2] = 0;
A[8, 0] = -0.0251625000000786; A[8, 1] = -35.9106333333333; A[8, 2] = 0;
A[9, 0] = -0.026762500000018; A[9, 1] = -39.7074333333333; A[9, 2] = 0;
A[10, 0] = -0.0261625000000549; A[10, 1] = -43.1843333333333; A[10, 2] = 0;
A[11, 0] = -0.010362500000042; A[11, 1] = -53.1814333333333; A[11, 2] = 0;
A[12, 0] = 0.0203374999999824; A[12, 1] = -62.1617333333333; A[12, 2] = 0;
A[13, 0] = 0.0272374999999556; A[13, 1] = -63.9696333333333; A[13, 2] = 0;
A[14, 0] = -0.0147625000000744; A[14, 1] = -72.0996333333333; A[14, 2] = 0;
A[15, 0] = -0.0130625000000464; A[15, 1] = -69.6783333333333; A[15, 2] = 0;
A[16, 0] = -0.00576250000005984; A[16, 1] = -67.0790333333333; A[16, 2] = 0;
A[17, 0] = 0.00063749999992524; A[17, 1] = -0.530333333333337; A[17, 2] = 0;
A[18, 0] = -0.000762500000064392; A[18, 1] = 4.58916666666666; A[18, 2] = 0;
A[19, 0] = 0.0177374999999529; A[19, 1] = 27.9494666666667; A[19, 2] = 0;
A[20, 0] = 0.0226374999999734; A[20, 1] = 31.8685666666667; A[20, 2] = 0;
A[21, 0] = 0.00643749999994725; A[21, 1] = 81.9918666666667; A[21, 2] = 0;
A[22, 0] = 0.00663750000001073; A[22, 1] = 86.5695666666667; A[22, 2] = 0;
A[23, 0] = 0.00853749999998854; A[23, 1] = 88.8083666666667; A[23, 2] = 0;If I let Math.Net Numerics do the Svd and output the VT matrix I get the following result:
-0.000077323216972426711 -0.99999999701056008 0.0 -0.99999999701056008 0.000077323216972426711 0.0 0.0 0.0 1.0
If I use the netlib implementation I get the following for VT:
-0.000077323216972426589 -0.99999999701056019 0.0 0.99999999701056019 -0.000077323216972426589 0.0 0.0 0.0 1.0
Also the U matrix is different but this is too much to post here...
If I remember correctly, the managed Math.NET code is a port of the LINPACK SVD routine. I don't know the differences between the LAPACK and LINPACK implementations. The native provider uses LAPACK's DGESVD. Is the U matrix as close as the VT matrix (which seems to be the same within 14 decimal places)?
Regards,
Marcus
The U matrix is as close as the VT matrix if you look at the absolute values, but the first column of the U matrix is negated.
Similar is the VT matrix but here it is the second row which is negated. As I know now, that Math.NET uses LINPACK SVD instead of LAPACK's DGESVD routine I'll try to find something with google. I thought Math.NET uses LAPACK's DGESVD and was wondering why the results are different..
Thanks for that Info!
Hi,
every DenseEvd seems to throw an error if the matrix is not square, even though Hessenberg and real Schur conversions seem to be implemented. There is also a bool that is checked to see if the input matrix is symmetric, but it is set to true in the evd code itself. I do not know if this is intended. If it is not, i may open an issue.
Kekze
Ok after looking at Math.Net Numerics code I found the comment, that it's SVD is equivalent to the GESVD LAPACK routine.
What is the difference between LAPACK GESVD and LAPACK DGESVD? I know this is maybe a stupid question but I couldn't find something with google and I'm still wondering why I get different results (in case GESVD and DGESVD are the same).
Thanks
>that it's SVD is equivalent to the GESVD LAPACK routine.
The comment was meant for the ILinearAlgebra interface and probably shouldn't have been copied to the managed code implementation. I'll remove it in both places to prevent future confusion.
>What is the difference between LAPACK GESVD and LAPACK DGESVD?
DGESVD is the double precision version of GESVD, SGESVD is the single precision, and so on.
SVD is only unique up to to the sign (for real numbers, I think it is a little more complicated for complex numbers). From a quick search it seems that LAPACK uses QR decomp to compute the SVD while LINPACK uses Lanczos iterations. That probably explains the
differences in signs.
For a given function f: x->f(x) I would like to compute numericallyits inverse function.
In other word for a given y, I'd like to compute x such f(x) = y.
Is there anything in the library to do so?
Well that could explain it. Thanks for your help and sorry for some maybe stupid questions..
I at least should have been able to answer the GESVD and DGESVD myself..
double x = SpecialFunctions.Factorial(14); // x will now have the value 87178291200.0 double y = SpecialFunctions.Factorial(31); // y will now have the value 8.2228386541779224E+33
GAMMA(a) = int(exp(-t)t^(a-1), t=0..infinity) gamma(a,x) = int(exp(-t)t^(a-1),t=0..x) Gamma(a,x) = int(exp(-t)t^(a-1),t=x..infinity) P(a,x) = gamma(a,x)/GAMMA(a) Q(a,x) = Gamma(a,x)/GAMMA(a)
double x = 0.9; double res = SpecialFunctions.Erf(x); // res will now have the value 0.7969082124