MTL: Loop-free submatrix extraction

[Michael Smolsky-2]

Peter,

Thank you for your detailed response.

I'm exploring MTL further. I hope you don't mind if I ask you some questions, the answers to which I couldn't find in the documentation.

I need to extract a submatrix of a matrix, in a way similar to:

octave:4> a=rand(4),b=a([1,2],[2,4])
a =

   0.622575   0.844744   0.424293   0.207615
   0.642011   0.102342   0.011919   0.168107
   0.774028   0.446422   0.076552   0.759136
   0.633285   0.792054   0.738877   0.982645

b =

   0.84474   0.20762
   0.10234   0.16811

Is there an elegant way of doing this, or I need to write loops? I feel optimistic having read about inserters in your documentation...

My matrix is dense for now, but I will probably need to apply this to a sparse matrix as well.

Thanks you in advance,

Michael.

> ----- Original Message -----
> From: "Peter Gottschling" <pgottsch@...>
> To: "General Matrix Template Library (MTL) list" <mtl-devel@...>, "Matrix Template Library 4" <mtl4@...>
> Subject: Re: MTL: MTL development road
> Date: Mon, 03 Dec 2007 10:53:38 +0100
>
>

--
Want an e-mail address like mine?
Get a free e-mail account today at www.mail.com!


_______________________________________________
This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/

[Peter Gottschling]

Hi again, Michael,

For dense2D and morton_dense there is a function sub_matrix (shame on me that there is no documentation yet).  You can use it like:

b= sub_matrix(a, begin_row, end_row, begin_column, end_column);

As you see it only supports intervals of rows and columns not arbitrary sets.  The reason is that this can be realized efficiently with a view.

If you are going to write a function I would do something like (only sketch not tested code):

template <typename Matrix, typename RowVector, typename ColVector>

Matrix list_sub_matrix(const Matrix& matrix, const RowVector& row_vector, const ColVector& col_vector)

{

  Matrix s(size(row_vector), size(col_vector));

  {

    matrix::inserter<Matrix>   ins(s);

    for (size_t r= 0; r < size(row_vector); r++)

       for (size_t c= 0; c < size(col_vector); c++)

            ins(r, c) << matrix[row_vector[r]][col_vector[c]];

  }

  return s;

}

Returning a matrix is quite expensive at the moment but will be okay once I've implemented move semantics.  I hope I haven't overseen a detail otherwise you can asked again.  Once you have it done and tested I'd like to put it in the repository.

Good luck and Merry Christmas,

Peter

On Dec 22, 2007, at 8:24 AM, Michael Smolsky wrote:

Peter,

Thank you for your detailed response.

I'm exploring MTL further. I hope you don't mind if I ask you some questions, the answers to which I couldn't find in the documentation.

I need to extract a submatrix of a matrix, in a way similar to:

octave:4> a=rand(4),b=a([1,2],[2,4])

a =

   0.622575   0.844744   0.424293   0.207615

   0.642011   0.102342   0.011919   0.168107

   0.774028   0.446422   0.076552   0.759136

   0.633285   0.792054   0.738877   0.982645

b =

   0.84474   0.20762

   0.10234   0.16811

Is there an elegant way of doing this, or I need to write loops? I feel optimistic having read about inserters in your documentation...

My matrix is dense for now, but I will probably need to apply this to a sparse matrix as well.

Thanks you in advance,

Michael.

----- Original Message -----

From: "Peter Gottschling" <pgottsch@...>

To: "General Matrix Template Library (MTL) list" <mtl-devel@...>, "Matrix Template Library 4" <mtl4@...>

Subject: Re: MTL: MTL development road

Date: Mon, 03 Dec 2007 10:53:38 +0100

-- 

Want an e-mail address like mine?

Get a free e-mail account today at www.mail.com!

_______________________________________________

This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/

------------

Peter Gottschling

Research Associate

Open Systems Laboratory

Indiana University

135 Lindley Hall

Bloomington, IN 47405

Tel.: [phone]   Fax: [phone]

http://www.osl.iu.edu/~pgottsch

_______________________________________________
This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/

[Michael Smolsky-2]

Peter,

I seem to be unable to say

result += op1 * transposed_view <const M> (op2),

where op1, op2, result are of the type dense2D<double> and M is a typedef for dense2D<double>.

The compiler barks at the absense of a type named type in struct mtl::traits::mult_result<compressed2D<...>, mtl::transposed_view<compressed2D<..> > >.

The code, that tests a product of a dense and a transposed matrix seems to be commented out from matrix_product_test.cpp.

Does MTL support calculation of the product of a transposed and dense matrix?

Thank you,

Michael

--

Got No Time? Shop Online for Great Gift Ideas!
mail.com Shopping

_______________________________________________
This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/

[Peter Gottschling]

Hi Michael,

Thank you for reminding me on that.  I thought I already fixed that.  Early January I hope finding the time for that.

Cheers,

Peter

On Dec 26, 2007, at 8:44 PM, Michael Smolsky wrote:

Peter,

I seem to be unable to say

result += op1 * transposed_view <const M> (op2),

where op1, op2, result are of the type dense2D<double> and M is a typedef for dense2D<double>.

The compiler barks at the absense of a type named type in struct mtl::traits::mult_result<compressed2D<...>, mtl::transposed_view<compressed2D<..> > >.

The code, that tests a product of a dense and a transposed matrix seems to be commented out from matrix_product_test.cpp.

Does MTL support calculation of the product of a transposed and dense matrix?

Thank you,

Michael

--

Got No Time? Shop Online for Great Gift Ideas!
mail.com Shopping

_______________________________________________

This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/

------------

Peter Gottschling

Research Associate

Open Systems Laboratory

Indiana University

135 Lindley Hall

Bloomington, IN 47405

Tel.: [phone]   Fax: [phone]

http://www.osl.iu.edu/~pgottsch

_______________________________________________
This list is archived at http://www.osl.iu.edu/MailArchives/mtl-devel/