Does anybody know how I can load parts of a matrix into matlab instead of all at once? because matlab won't allow me to load it all at once because the matrix is too big. Thanks.
Subject: breaking up a large matrix before loading into matlab
gary12345 wrote:
> Does anybody know how I can load parts of a matrix into matlab
> instead of all at once? because matlab won't allow me to load it all
> at once because the matrix is too big. Thanks.
If it's a binary or text file see the doc for the appropriate input
function--they have optional arguments to include number of elements to
read.
If some other ML system did a SAVE to a .MAT file, there I'm not sure
outside a mex file or other hack to get at the data around the structure
of the file that there is...
--
Subject: breaking up a large matrix before loading into matlab
"Sprinceana " <mihaispr@yahoo.com> wrote in message <h2k8dh$6uc$1@fred.mathworks.com>...
> For example if your big matrix (stored in variable) is 720x960 and you want to load only 10*10:
>
>
> img=imread('poza.jpg');imshow(img);
>
>
> >>whos img
>
> Name Size Bytes Class
>
> img 720x960 691200 uint8 array
>
> Make this (to store only 10*10 part of that matrix):
>
> small_img=img(1:10,1:10);
> imagesc(small_img);
SO... you end up having the BIG -AND- the SMALL matrix in your workspace...
how does this help the OP...
us
Subject: breaking up a large matrix before loading into matlab
gary12345 <garycyhou@yahoo.ca> wrote in message <18789979.66027.1246568708618.JavaMail.jakarta@nitrogen.mathforum.org>...
> Does anybody know how I can load parts of a matrix into matlab instead of all at once? because matlab won't allow me to load it all at once because the matrix is too big. Thanks.
a hint:
help memmapfile;
us
Subject: breaking up a large matrix before loading into matlab
dpb <none@non.net> wrote in message <h2jh0c$edq$2@news.eternal-september.org>...
> gary12345 wrote:
> > Does anybody know how I can load parts of a matrix into matlab
> > instead of all at once? because matlab won't allow me to load it all
> > at once because the matrix is too big. Thanks.
>
> If it's a binary or text file see the doc for the appropriate input
> function--they have optional arguments to include number of elements to
> read.
>
> If some other ML system did a SAVE to a .MAT file, there I'm not sure
> outside a mex file or other hack to get at the data around the structure
> of the file that there is...
I have not seen he internal format of MAT file documented anywhere. Not sure how to hack it.
Best recommendation is to find a computer with Matlab and large RAM to break down the file.
Few possible attempts: Try first to load the file after CLEAR ALL command. It is possible that the SAVE command is carried out on the same computer with large data-sharing variables. LOAD - since unshared them - thus will be unable to read it on the very same computer. In this case load individual variables separately - interleaved with clear all - might work. If it doesn't pass, then you are screwed.
Bruno
Subject: breaking up a large matrix before loading into matlab
"Bruno Luong" <b.luong@fogale.findmycountry> wrote in message <h2k9j1$l96$1@fred.mathworks.com>...
> dpb <none@non.net> wrote in message <h2jh0c$edq$2@news.eternal-september.org>...
> > gary12345 wrote:
> > > Does anybody know how I can load parts of a matrix into matlab
> > > instead of all at once? because matlab won't allow me to load it all
> > > at once because the matrix is too big. Thanks.
> >
> > If it's a binary or text file see the doc for the appropriate input
> > function--they have optional arguments to include number of elements to
> > read.
> >
> > If some other ML system did a SAVE to a .MAT file, there I'm not sure
> > outside a mex file or other hack to get at the data around the structure
> > of the file that there is...
>
> I have not seen he internal format of MAT file documented anywhere. Not sure how to hack it.
>
> Best recommendation is to find a computer with Matlab and large RAM to break down the file.
>
> Few possible attempts: Try first to load the file after CLEAR ALL command. It is possible that the SAVE command is carried out on the same computer with large data-sharing variables. LOAD - since unshared them - thus will be unable to read it on the very same computer. In this case load individual variables separately - interleaved with clear all - might work. If it doesn't pass, then you are screwed.
>
> Bruno
hm... i didn't think the OP was talking about a MAT-file...
anyhow, here's the doc describing the anatomy of a MAT...
On 2 Jul, 23:04, gary12345 <garycy...@yahoo.ca> wrote:
> Does anybody know how I can load parts of a matrix into matlab instead of all at once? because matlab won't allow me to load it all at once because the matrix is too big. Thanks.
If you have a wintel multi-core PC, then try this:
- Push <ctrl>-<alt>-<del> to bring up the task manager
- Right-click on the matlab process in the processes view
- Select 'set affinity...' in the pop-up menu
- De-select CPU0
- Press 'OK'
This will cause the matlab process to be run on one
of the other cores available. With a little luck,
the OS and most of the other processes run on CPU0
core, meaning lots of the memory associated with this
core is not available to matlab.
If this doesn't work, try the same again but deselect
some other core.
NOTICE: Any content you submit to MATLAB Central, including personal information, is not subject to the protections which may be afforded information collected under other sections of The MathWorks, Inc. Web site. You are entirely responsible for
all content that you upload, post, e-mail, transmit or otherwise make available via MATLAB Central. The MathWorks does not control the content posted by visitors to MATLAB Central and, does not guarantee the accuracy, integrity, or quality of such content.
Under no circumstances will The MathWorks be liable in any way for any content not authored by The MathWorks, or any loss or damage of any kind incurred as a result of the use of any content posted, e-mailed, transmitted or otherwise made available
via MATLAB Central.
Read the complete Terms prior to use.
Feed for this Thread