Re: reading HDF5 slices in octave

[David Bateman]

Nuno Nunes wrote:

> On Sat, 2005-11-05 at 10:06 +0100, David Bateman wrote:
>  
>
> > Well "load <file> <var>" should load the variable <var> from <file>. In 
> > the case of hdf5 files the group name is assumed to be the variable 
> > name. So I think we can load single variables from an hdf5 file. As for 
> > NDArray stuff as Quentin pointed out it was added in 2.1.53.
> >    
>
> I know about the "load <file> <var>" syntax for loading selected
> variables only, but not of any way of reading a 2D slice of a 3D
> variable, which is what I expected to use h5read.oct for. I'm working
> with 1735*4320*66 vars, and would like to load say a 4320*66 slice. Does
> octave's load command have this capability? If so, it doesn't seem to be
> documented anywhere. What is the option syntax?
>
> If that's not the case, I think there would be good reason to re-include
> the h5read plugin in h5utils. Even if I probably don't REALLY need it,
> now that I've had time to think of alternative ways of solving my
> problem...
>
> Many thanks for all your comments,
> Nuno
>
>  
Even the old octave HDF5 didn't have this capability. How much memory do 
you have? I must admit your matrix is large and so you'd need 4GB, but 
just loading the whole variable as an NDArray and slicing it up later 
will work.. IE.

load <file> myvar
sliceN = myvar(N,:,:);

D.

--
David Bateman                                address@hidden
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 1 69 35 77 01 (Fax) 
91193 Gif-Sur-Yvette FRANCE

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------