text_classification

Why we included this project

This repo collects a wide spread of classic deep learning models for text classification, from fastText and TextCNN up through attention-based seq2seq, transformers, and memory networks. The breadth is the point: you can run several very different architectures on the same data and see which one suits your task instead of betting on a single approach up front. Multi-label classification is supported, which covers real-world jobs like tagging documents or customer tickets with several labels at once. Every model has its own test function and is independent of the dataset, so you can sanity-check it on a toy task before pointing it at your own data. One caveat: the code targets older TensorFlow and Python, so expect to port things if you're on a current stack.