Thread Subject:

Griding data

Dear Friends, I am facing problem in Griding data in matlab. I have written the code but its taking sooo much time to process. My data contains X(longitude) Y(latitude) Z(data points) with 0.5 degree resolution. I have to make this data gridding with 1.5x1.5 degree resolution. please help me how to save the time and also i have gone through all the griding programs in mathworks.com.

On 4/3/2012 10:14 PM, Shaik Ghouse Basha wrote:
> Dear Friends, I am facing problem in Griding data in matlab.
>I have written the code but its taking sooo much time to process.
>My data contains X(longitude) Y(latitude) Z(data points) with 0.5 degree
>resolution. I have to make this data gridding with 1.5x1.5 degree
> resolution. please help me how to save the time and also i have gone
>through all the griding programs in mathworks.com.

It will help others help you more by making small example
of the problem, post the complete code of the small example
here.

Hard to help you without seeing anything.

--Nasser

I have X Y Z of length 13000 like this i have 16700 columns of data
for j=1
eps1=1.5;
        Eda=nan(length(lat1),length(lon1));
        for k=1:length(Z)
            
            P1=find(lat1>=Y(k)-eps1 & lat1<=Y(k)+eps1);
            Q1=find(lon1>=X(k)-eps1 & lon1<=X(k)+eps1);
            if length(P1)>1
                P1=P1(1);
            end
            if length(Q1)>1
                Q1=Q1(1);
            end
            
            Eda(P1,Q1)=nanmean([Eda(P1,Q1) Z(k)]);
        end
end

"Nasser M. Abbasi" <nma@12000.org> wrote in message <jlgfdf$ujr$2@speranza.aioe.org>...
> On 4/3/2012 10:14 PM, Shaik Ghouse Basha wrote:
> > Dear Friends, I am facing problem in Griding data in matlab.
> >I have written the code but its taking sooo much time to process.
> >My data contains X(longitude) Y(latitude) Z(data points) with 0.5 degree
> >resolution. I have to make this data gridding with 1.5x1.5 degree
> > resolution. please help me how to save the time and also i have gone
> >through all the griding programs in mathworks.com.
>
> It will help others help you more by making small example
> of the problem, post the complete code of the small example
> here.
>
> Hard to help you without seeing anything.
>
> --Nasser
>

"Shaik Ghouse Basha" wrote in message <jlggfu$1ua$1@newscl01ah.mathworks.com>...
> I have X Y Z of length 13000 like this i have 16700 columns of data
> for j=1
> eps1=1.5;
> Eda=nan(length(lat1),length(lon1));
> for k=1:length(Z)
>
> P1=find(lat1>=Y(k)-eps1 & lat1<=Y(k)+eps1);
> Q1=find(lon1>=X(k)-eps1 & lon1<=X(k)+eps1);
> if length(P1)>1
> P1=P1(1);
> end
> if length(Q1)>1
> Q1=Q1(1);
> end
>
> Eda(P1,Q1)=nanmean([Eda(P1,Q1) Z(k)]);
> end
> end
>
> "Nasser M. Abbasi" <nma@12000.org> wrote in message <jlgfdf$ujr$2@speranza.aioe.org>...
> > On 4/3/2012 10:14 PM, Shaik Ghouse Basha wrote:
> > > Dear Friends, I am facing problem in Griding data in matlab.
> > >I have written the code but its taking sooo much time to process.
> > >My data contains X(longitude) Y(latitude) Z(data points) with 0.5 degree
> > >resolution. I have to make this data gridding with 1.5x1.5 degree
> > > resolution. please help me how to save the time and also i have gone
> > >through all the griding programs in mathworks.com.
> >
> > It will help others help you more by making small example
> > of the problem, post the complete code of the small example
> > here.
> >
> > Hard to help you without seeing anything.
> >
> > --Nasser
> >

you might find this function useful:

http://www.mathworks.com/matlabcentral/fileexchange/23479-bin2mat

It is not optimized for speed, but may be faster than your current code.
-Andrew

On Wednesday, April 4, 2012 3:51:26 PM UTC+12, Shaik Ghouse Basha wrote:
> I have X Y Z of length 13000 like this i have 16700 columns of data
> for j=1
> eps1=1.5;
> Eda=nan(length(lat1),length(lon1));
> for k=1:length(Z)
>
> P1=find(lat1>=Y(k)-eps1 & lat1<=Y(k)+eps1);
> Q1=find(lon1>=X(k)-eps1 & lon1<=X(k)+eps1);
> if length(P1)>1
> P1=P1(1);
> end
> if length(Q1)>1
> Q1=Q1(1);
> end
>
> Eda(P1,Q1)=nanmean([Eda(P1,Q1) Z(k)]);
> end
> end
>
> "Nasser M. Abbasi" <nma@12000.org> wrote in message <jlgfdf$ujr$2@speranza.aioe.org>...
> > On 4/3/2012 10:14 PM, Shaik Ghouse Basha wrote:
> > > Dear Friends, I am facing problem in Griding data in matlab.
> > >I have written the code but its taking sooo much time to process.
> > >My data contains X(longitude) Y(latitude) Z(data points) with 0.5 degree
> > >resolution. I have to make this data gridding with 1.5x1.5 degree
> > > resolution. please help me how to save the time and also i have gone
> > >through all the griding programs in mathworks.com.
> >
> > It will help others help you more by making small example
> > of the problem, post the complete code of the small example
> > here.
> >
> > Hard to help you without seeing anything.
> >
> > --Nasser
> >

Well, I can see why it's taking so long.
It's those find commands.
They assume your data are randomly distributed, but actually they're in a grid aren't they?
In which case, you can reshape the vectors into matrices and apply a bit of intelligence to how you process the data.
What are the dimensions of your grid?