Regarding fval in Decision Tree

#1

i want to ask whether fval for any node in the tree will be a clever choice if we took labels.mean() for any type of categorical data

#2

hey @amankharb ,
yes you are correct , its not a clever choice to take mean of any type categorical data.
In the video it just shown for understanding purpose , as there are many other parameters too in decision trees which requires an in depth knowledge over this topic.

For categorical data, we can various statistic methods to find the class which is nearest to mean value,or we use mode or any other method. Its just our choice the way we implement it and use it.
How it effects the model performance and learning is the main objective.

So it is just to understand an overview , of how it works. To learn more about in depth , search it over google and understand its code on github.

I hope this would have resolved your doubt.
Thank You and Happy Learning :slightly_smiling_face:.

#3

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.