Did you normalize the dataset?

Leen Faris Chances are that you are using 0 and 1 as real numbers for training and not categorical data. If you use categorical labels for training in MATLAB, only then your outputs would be categorical. Please check if you can follow these example codes using categorical data for classification to get your job done

https://www.mathworks.com/help/vision/ug/image-category-classification-using-deep-learning.html

You are probably using regression output layer in your code: https://www.mathworks.com/help/deeplearning/ref/regressionlayer.html

You should use a classification output layer instead:

https://www.mathworks.com/help/deeplearning/ref/classificationlayer.html

But I guess the easiest way for you would be to use some classification code that I referred to above, and just change the training sample and their labels to suit your task. If needed, you can change real number training labels (0 or 1 as numbers) to categorical labels (0 and 1 as classes) using the command

categorical

https://www.mathworks.com/help/matlab/ref/categorical.html

The following link can be useful for you:

https://www.mathworks.com

Did you normalize the dataset?

Mohammed Ashikur Rahman of course, I normalized data with minmax scalar then trained anfis, also normalized data with z-score scalar and trained anfi. and the same result in tow methods