why use sparse matrices in matlab?
in [Matlab]
|
From: shahnaz fatima on 28 Jun 2010 02:51 i ahave a doubt. why use sparse matrices in matlab. i think when we find sparse of a matrix some data is lost . can we do it reverse? is there any function to desparse?. is sparse matrices used only denote the nodes of graphs like that? kindly reply
From: us on 28 Jun 2010 03:00 "shahnaz fatima" <shahnaz1981fat(a)gmail.com> wrote in message <i09gop$g0n$1(a)fred.mathworks.com>... > i ahave a doubt. why use sparse matrices in matlab. > i think when we find sparse of a matrix some data is lost . can we do it reverse? is there any function to desparse?. > > is sparse matrices used only denote the nodes of graphs like that? > > kindly reply a hint: - you will find plenty of information here... http://www.mathworks.com/access/helpdesk/help/techdoc/ref/f16-5872.html#f16-6492 us
From: James Tursa on 28 Jun 2010 03:02 "shahnaz fatima" <shahnaz1981fat(a)gmail.com> wrote in message <i09gop$g0n$1(a)fred.mathworks.com>... > > i ahave a doubt. why use sparse matrices in matlab. To save memory, and sometimes to save time. > i think when we find sparse of a matrix some data is lost Yes, the zeros are lost ... but not forgotten. > . can we do it reverse? is there any function to desparse?. full > is sparse matrices used only denote the nodes of graphs like that? No. James Tursa
From: shahnaz fatima on 28 Jun 2010 07:20 hello james tursa. thank you for the point to point reply -- shahnaz
From: John D'Errico on 28 Jun 2010 09:25 "shahnaz fatima" <shahnaz1981fat(a)gmail.com> wrote in message <i09gop$g0n$1(a)fred.mathworks.com>... > i ahave a doubt. why use sparse matrices in matlab. > i think when we find sparse of a matrix some data is lost . can we do it reverse? is there any function to desparse?. > > is sparse matrices used only denote the nodes of graphs like that? There are many things that people never use. What is the point of those features? Clearly none. In fact, anything that I never use is completely useless. Why did they waste any time writing it at all? Of course, all of that is false. In your case, you probably never have a use for large scale linear algebra. Or, possibly you don't know that you do. Perhaps you have never had the need to use tools like my inpaint_nans or gridfit from the file exchange. These are tools that absolutely need sparse matrices to run in an essentially finite amount of time. If you solve large finite element problems, large scale optimization problems, or many others, you will need to use sparse matrices. John
|
Pages: 1 Prev: Opening a Image in MatLAB GUI Next: Listener to HG callback |