Why do I receive 'You must pass class labels as a vector.' error?

clear; clc; 
X_train = readmatrix("x_train.csv",'NumHeaderLines',1);
X_test = readmatrix("x_test.csv",'NumHeaderLines',1);
y_train = readmatrix("y_train.csv",'NumHeaderLines',1);
y_test = readmatrix("y_test.csv",'NumHeaderLines',1);

model = fitcsvm(X_train, y_train, 'KernelFunction','rbf');

Here's my code. After trying to run I receive this error (error is in attached screenshot):

Answer 1

[Hiro Yoshino]

It said the y_train is not a vector.

The argument should be any on of: categorical array, string array, character array, logical array, numerical array, cell array.

https://jp.mathworks.com/help/stats/fitcsvm.html?s_tid=doc_ta#bt9w6j6-Y

You should make sure you give right data for x and y respectively.

[Daniil Podtesov]

Thanks for the answer, y_train contains numercial and boolean values, is it considered as a numerical?

[Hiro Yoshino]

How could Y be two values? It should be one. (This must be a classification problem.)

You should extract either one and feed it into fitcsvm as appropriate.

[Daniil Podtesov]

I see, in Python Y was two cols (id, boolean), that's why I remained this column. Thank You, I'll try the solution

[Hiro Yoshino]

Good luck 🤞