This repo contains the implementation of the paper 'Word Embedding Perturbation for Sentence Classification'

It also supports basic CNN versions of sentiment classification, relation classification and answer selections.

First, you need to download 300-dim pretrained word2vec embeddings https://code.google.com/archive/p/word2vec/ https://nlp.stanford.edu/projects/glove/

All the methods are implemented using Tensorflow v1.0.1.