help with using surface fit function

I cant figure out whats wrong - but i just cat get this function to work. I have a 2D array of values (256x256) - z

I define X1 as 1x256 vector

Y1 as 1x256 vector

trying to use this, according to an example from the Help Docs:

sf=fit ( [X1,Y1],z,'Exclude', z < 10)

i get abunch of errors which I cant even understand.

Error using fit>iParseOptionalArgs

Algorithm options must be specified in a FITOPTIONS object or as property-value pairs.

Error in fit (line 113)

[useroptions, useroptargs, probparams] = iParseOptionalArgs( varargin{:} );

Answer 1

[Matt J]

You haven't specified the fit type. Also, your X, Y, and z elements all need to be (256^2)x1 vectors.

[XX,YY]=ndgrid(X1,Y1)

sf=fit ( [XX(:),YY(:)],z(:),fitType, 'Exclude', z < 10)

[Alexander Lazarev]

Thanks Matt, i need clarification 256 squared? i dont understand..

here is what i tried just now:

X1 is a column vector 256x1 (values are from 1, 2, 3, ... to 256)

Y1 is a row vector 1x256 (values are from 1, 2, 3, ... to 256)

>> sf=fit([X1,Y1],z,'linearinterp','Exclude', z < 100)

Error using horzcat

Dimensions of arrays being concatenated are not consistent.

please help me gt this thing to work )))))

[Matt J]

What happened when you tried what I posted? X,Y, and Z all need to be column vectors of the same length.

[Alexander Lazarev]

>> sf=fit ( [XX(:),YY(:)],z(:),'linearinterp', 'Exclude', z < 10)

Error using curvefit.interpoptions/set

Matrix must be column or row vector.

Error in fitoptions (line 167)

set(f,varargin{:}); % should not allow method to be set.

Error in fit>iGetFitOptions (line 1169)

options = fitoptions(options, useroptargs{:});

Error in fit>iFit (line 178)

options = iGetFitOptions( model, useroptions, useroptargs, warningFcn );

Error in fit (line 116)

[fitobj, goodness, output, convmsg] = iFit( xdatain, ydatain, fittypeobj, ...

[Alexander Lazarev]

my data array has 256x256 matrix of values...

I am attaching the image of my data here.

I am trying to work on fitting this data with a surface..

[Alexander Lazarev]

repeating this for convenience. Here is what i get if i use the sintax you suggested. (i created XX anf YY per your sintax as well)

>> sf=fit ( [XX(:),YY(:)],z(:),'linearinterp', 'Exclude', z < 10)

Error using curvefit.interpoptions/set

Matrix must be column or row vector.

Error in fitoptions (line 167)

set(f,varargin{:}); % should not allow method to be set.

Error in fit>iGetFitOptions (line 1169)

options = fitoptions(options, useroptargs{:});

Error in fit>iFit (line 178)

options = iGetFitOptions( model, useroptions, useroptargs, warningFcn );

Error in fit (line 116)

[fitobj, goodness, output, convmsg] = iFit( xdatain, ydatain, fittypeobj, ...

[Matt J]

The Exclude option must also be given input in vector form. From the doc:

Exclude — Points to exclude from fit

expression | index vector | logical vector | empty

sf=fit ( [XX(:),YY(:)],z(:),fitType, 'Exclude', z(:) < 10)

[Alexander Lazarev]

thank you it finally worked )))