Hi there, we would like to start using XGBoost for a Classification problem.
The question is: all features and targets has to be numerical or also alphanumerical?
Our dataset is a list of warehouse items: there are descriptions, codes, numbers…etc.
Are we supposed to pre-process all non-numerical data in numbers (ex. word embeddings for descriptions) or XGBoost handles by design all data types?
Thank’s in advance
