Implementation of Traditional Vs. Transformer Machine Learning Models

Abstract

Classical models have been mostly used in NLP tasks. There are several classical models like Naïve Bayes, SVM, CNN etc. In classical model we need to build the model from scratch. The model is trained, developed and then lastly deployed. With the advent of transformer models the process have been simplified. Transformer models are pretrained on larger datasets and just needs labelled dataset. It saves time, energy and resources. This report discusses the mechanism involved in workflow of classical and transformer ml models. Implementation of models have been also done for comparison.