[MATRIX-SIG] A proposal (LONG) - "reverse of take" revisited

[Konrad Hinsen]

> I would like to see block capability added to NumPy. Memory management,

Much of it is already there! You just have to play around with shapes.

Example: I frequently work with matrices that logically consist of 3x3
blocks. For some operations I must work on the blocks, for others on
the whole matrix. So here we go:

  matrix_of_blocks = zeros((n, 3, n, 3), Float)
  big_matrix = reshape(matrix_of_blocks, (3*n, 3*n))

Since reshape() returns a reference sharing the same data space,
I can use the two arrays interchangeably.

The only restriction is that the blocks must be of equal size.

> * A slice that does not make a copy, so I can modify a part of an image.

That's what indexing already does!

> * Efficiently view a big array as an array of blocks (each an array
> itself) and vice versa.
> * Convert an array of arrays to and from a big array.

Possible with restrictions (see above).

> * The ability to extract, insert, and manipulate sub-blocks of blocks.

Within the same restrictions, simply done by indexing.

Question: do you see a need for supporting non-equal-sized blocks?
That could become very messy.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                          | E-Mail: hinsen@ibs.ibs.fr
Laboratoire de Dynamique Moleculaire   | Tel.: +33-4.76.88.99.28
Institut de Biologie Structurale       | Fax:  +33-4.76.88.54.94
41, av. des Martyrs                    | Deutsch/Esperanto/English/
38027 Grenoble Cedex 1, France         | Nederlands/Francais
-------------------------------------------------------------------------------

_______________
MATRIX-SIG  - SIG on Matrix Math for Python

send messages to: matrix-sig@python.org
administrivia to: matrix-sig-request@python.org
_______________