[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Using otpiminterp in octave

From: Doug Stewart
Subject: Re: Using otpiminterp in octave
Date: Tue, 7 Oct 2014 17:47:00 -0400

On Tue, Oct 7, 2014 at 5:37 PM, Zilore Mumba <address@hidden> wrote:
Further help sought. With the kind help I got so far I do get the correct interpolated values.
Being a complete beginner in octave, I have been going through documentation and tried several ways of writing formatted output without success. I would like my output to be printed as x, y, z

%result of interpolatin
obsi = fi + mean_obs;

% put x, y, z into a matrix beofre printing
matxy = [xi yi obsi];

fid = fopen('mydata.txt','w');
fprintf(fid, '%5.2f %5.1f %5.1f\n', xi,yi,obsi);

%for j = 1:length(yi),
%   for i = 1:length(xi),
%       fprintf(fid, '%5.2f %5.1f %5.1f\n', matxy);
%   end

try this
   for i = 1:length(xi),
      fprintf(fid, '%5.2f %5.1f %5.1f\n', xi(i),yi(i),obsi(i) );

The two fprintf statements give me all x, followed by all y then all z values.
The for loop gives me only 121 values, the length of the individual vectors x, y, z.

I know after a long time of trying and browsing documentation, I may get. But to save time I would appreciate someone's time to help me out.

On Tue, Oct 7, 2014 at 6:12 PM, Zilore Mumba <address@hidden> wrote:
Thanks very much, I did not see that silly slip I made. Now I get the results I expect. The assistance is highly appreciated.

On Tue, Oct 7, 2014 at 5:19 PM, Doug Stewart <address@hidden> wrote:

On Tue, Oct 7, 2014 at 10:55 AM, Zilore Mumba <address@hidden> wrote:
Thanks Ben, Doug for pointing me in the right direction. I get the error below. Initially I had put [xi,yi] = ndgrid((-22:0.1:34,-18:0.1:-8); When I removed the minus on 22 (i.e. [xi,yi] = ndgrid((22:0.1:34,-18:0.1:-8);) the error is pointing at 22.
While am still trying to see if I can see the right format in the documentation, Iwould appreciate further assistance.

octave:1> example_optiminterp_data
parse error near line 16 of file /home/zmumba/Optiminterp/example_optiminterp_data.m

  syntax error

>>> [xi,yi] = ndgrid((22:0.1:34,-18:0.1:-8);

You have 2 (( and only need 1 ( 

On Tue, Oct 7, 2014 at 2:46 PM, Doug Stewart <address@hidden> wrote:

On Tue, Oct 7, 2014 at 8:30 AM, Ben Abbott <address@hidden> wrote:

On Oct 7, 2014, at 8:09 AM, Zilore Mumba <address@hidden> wrote:

> The data I used is as in the attached file
> Just to be sure I read in the data as per lines below:
>Gold Coast Pipe Band
> % load the data file
> data=""> >
> % the 1st and 2nd column represent the location
> x = data(:,1);
> y = data(:,2);
> % the 3rd column represents the observations to interpolate
> obs = data(:,3);
> The output I get has no similarity with the input. Am not sure if am printing the error field.
> On Mon, Oct 6, 2014 at 11:21 PM, Doug Stewart <address@hidden> wrote:
> On Mon, Oct 6, 2014 at 1:01 PM, Zilore Mumba <address@hidden> wrote:
> I have just installed octave on ubuntu. Trying the example "example_optiminterp_data.m" with my own data, I get very strange values (very small values) see extract of output below
> # Created by Octave 3.8.1, Mon Oct 06 17:02:39 2014 CAT <address@hidden>
> # name: fi
> # type: matrix
> # rows: 41
> # columns: 41
>  -6.046732358407388e-147 -2.165934319425287e-147 -7.689701072432262e-148 -2.705909551864099e-148 -9.437495202085344e-149 -3.262419987996825e-149 -1.117796281402676e-149 -3.795990593671885e-150 -1.277695255495584e-150 -4.262546577144117e-151
> Here is the line I added for the output
> save mydata.txt fi;
> suspect am not printing the correct results.
> I apologise for this very basic question
> Help will be appreciated
> Zilore Mumba

The ranges for (x,y) and (xi,yi) do not overlap.

min (x)
ans =  23.100
ans =  33.200
min (y)
ans = -17.800
max (y)
ans = -8.9000

[xi,yi] = ndgrid(-2:0.1:2,-2:0.1:2);

should this be:
[xi,yi] = ndgrid(-22:0.1:34,-18:0.1:-8);

Try adding

 xi = median (x) + xi;
 yi = median (y) + yi;


DASCertificate for 206392

DASCertificate for 206392

DASCertificate for 206392

reply via email to

[Prev in Thread] Current Thread [Next in Thread]