Great repository! Thank you for your work!
I was wondering if it would be possible to extend the random forest classifier to deal with batched data. Instead of creating new trees in every call of fit, It would either update existing trees or add new trees based on some criterion like gini index or cross entropy. Is this a possible future work for this repo?
Great repository! Thank you for your work!
I was wondering if it would be possible to extend the random forest classifier to deal with batched data. Instead of creating new trees in every call of fit, It would either update existing trees or add new trees based on some criterion like gini index or cross entropy. Is this a possible future work for this repo?