Error using ~= Invalid types for comparison.

3 vues (au cours des 30 derniers jours)
Saraswathi S
Saraswathi S le 29 Juin 2021
Commenté : Saraswathi S le 2 Juil 2021
misclassrate = sum(Predicted ~= testdata.N)/numel(Predicted)
Here, my predicted variables are categorical while testdata.N is from a table.
Can anyone say what is the issue here?
  2 commentaires
KSSV
KSSV le 29 Juin 2021
Check the class of Predicted and testdata.N....they should be of same to compare.
Saraswathi S
Saraswathi S le 2 Juil 2021
Thank you so much. That was the issue. one was a categorical data and the other double.

Connectez-vous pour commenter.

Réponse acceptée

Walter Roberson
Walter Roberson le 29 Juin 2021
You probably used categorical(traindata.N) when you did the training -- that is, you probably took numeric data and converted it to categorical and used the categorical as the labels.
And how that it is time to validate, you have predicted categorical labels, but your testdata.N is still numeric. You have to convert it to categorical() in order to do the comparison.
misclassrate = sum(Predicted ~= categorical(testdata.N))/numel(Predicted)

Plus de réponses (0)

Catégories

En savoir plus sur Categorical Arrays dans Help Center et File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by