Info: https://www.kaggle.com/emilymu/cord19-analysis-scibert-embeddings-on-mesh-words
SQL_clean is the data-processing SQL pipeline forked from: https://github.com/neuml/cord19q
- Download/extract the CORD-19 Data to data
- Navigate to clustering folder
- Configure embedding script (You can modify build_embedding_functions.py)
- run python2.7 -m SimpleHttpServer
- Navigate to localhost:8000 and Open visualize.html