Predict the title of a materials science paper from its abstract using the magic of a bidirectional LSTM. To download/preprocess data (this takes a LONG time): ./download_and_preprocess.bash
Use abstract2title.py to train
Expects the following environment variables to be defined; MATSCHOLAR_STAGING_HOST MATSCHOLAR_STAGING_USER MATSCHOLAR_STAGING_PASS MATSCHOLAR_STAGING_DB