How to vary the "proportion" parameter - MulticlassOversampling class

[jnatmed]

Hello, I am trying to vary the proportion parameter in the MulticlassOversampling class:

I tried passing passing through the declaration of an instance of the class, but when executing it it had no effect.

I also tried in the declaration of the distance_SMOTE(proportion=0.5) method.

This is so, because the MulticlassOversampling class creates examples until it matches the majority class. However, I want to try different variants and not only create up to 100%, but 75 or 50%.
I hope I made myself understood.

[1ZhangMR]

Hello, have you solved it, I'm also modifying the portion parameter and it's not working!

[gykovacs]

Hi! Apologies for the late response. In the multi-class case, the proportion parameter is automatically determined for each class to match the majority class, so whatever you pass as proportion, will be overwritten. The reason for this is that in a multiclass case, the "proportion" is undefined: proportion of which class to which? Setting a proportion to 0.5 (in relation to the majority class) could lead to strange results, varying numbers of samples generated for each minority class, keeping the imbalance. Setting class level proportion parameters seems to me as a guarantee to overfitting, that is why it is not supported. I am 100% open to a discussion on the topic, and happy to add more controls to users, if it really makes sense.