Check out this excellent guide to text classification from Google. Covers important best practices and rules of thumb, including a complete flowchart for picking the right model & preprocessing pipeline based on the statistical structure of your dataset. developers.google.com/machine-learni… 
An important insight is that the ratio between number of training samples and mean number of words per sample can tell you whether you should be using a n-gram model or a sequence model -- and whether you should use pre-trained word embeddings or train your own from scratch.
This guide will serve as your free expert consultant for all your text classification problems. That flowchart is a form of AutoML! Of the cheap, effective kind.
