species2vec

bioRxiv pre-print: https://www.biorxiv.org/content/early/2018/11/05/461996

Usage instructions:

import gensim

# load species embeddings
m = gensim.models.KeyedVectors.load_word2vec_format('mammalia_6M.vec')

# test species embeddings
m.most_similar(u'Grammomys_ibeanus')

Two sets of embeddings are included in this repository (mammalia_6M.vec amd reptilia_3M.vec). If you need to generate new ones, have a look at the tutorial notebook in notebooks/species2vec_tutorial.ipynb.