An Extention of Dis2Vec model https://arxiv.org/pdf/1603.00106.pdf to documents

For a complete understanding of the work done here, take a look at https://slides.com/alexjohn-1/deck.

The entry point is the phraser_with_stemming.py file in thesis/gensim folder, this produces bigrams and trigrams used by the stem_pubmed.py file, which is the main training script.