I am working on a problem statement to develop a ML model to classify text. The problem i have here is to handle the high number of classes.. in my dataset, there are 2.5k diffe