Error: Unable to use a value of type 'fittype' as an index.

4 views (last 30 days)
Hi there,
I am struggling to understand how to solve an error occurring when using a customized non-linear fit function with Matlab.
Here is the function I need to employ:
CAO = 0.02;
expFIT = fittype(@(a, b, CD) a*((CA0+CD+b)-sqrt((CA0+CD+b).^2-4*CA0*CD)) ,'independent',{'CD'});
Up to here, no error.
Then, when it comes to perform the actual fit I get the following statement: '
Unable to use a value of type 'fittype' as an index'.
And it goes to the line with the fit command:
FIT = fit(CD, Abs, expFIT,'Robust','on');
Thank you in advance for any feedback!

Accepted Answer

Steven Lord
Steven Lord on 27 Nov 2018
Rename the variable you created with the name fit.
  3 Comments
Steven Lord
Steven Lord on 27 Nov 2018
Did you clear the variable named fit after you executed the command AbsFIT = fit; (or whatever similar command you used?)
Claudia Dessi
Claudia Dessi on 27 Nov 2018
No, I did not.
I did it this morning and now it works. It is curious though that FIT name for another fit command in another code was just fine, and I thought that capital letters may not create confusion to the program.
thanks!

Sign in to comment.

More Answers (1)

Prinkle Sharma
Prinkle Sharma on 10 Feb 2019
Edited: Prinkle Sharma on 10 Feb 2019
While compiling:
[~,score_knn] = resubPredict(KNNModel);
labels = predict(KNNModel,test(:,1:9));
I'm experiencing similar error: "Unable to use a value of type 'ClassificationKNN' as an index." with the line below
[Xknn, Yknn, Tknn, AUCknn] = perfcurve(labels,score_knn(:,KNNModel),'true');
Can someone help me with this please ?

Categories

Find more on Linear and Nonlinear Regression in Help Center and File Exchange

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!