I was able to train and use this on the COVID papers dataset

materialsintelligence / mat2vec

Supplementary Materials for Tshitoyan et al. "Unsupervised word embeddings capture latent knowledge from materials science literature", Nature (2019).

MIT License

616 stars 180 forks source link

I was able to train and use this on the COVID papers dataset #20

Closed Tylersuard closed 4 years ago

Tylersuard commented 4 years ago

I just wanted to say thank you :)

leotam commented 4 years ago

Do you have a link to your results?

jdagdelen commented 4 years ago

@Tylersuard I would love to hear more about what you're building with the embeddings you created.

Tylersuard commented 4 years ago

Indeed! https://www.kaggle.com/tylersuard/mat2vec-covid-papers-unexpected-word-asociations?scriptVersionId=30790274

I was just entering this for a Kaggle competition. They dumped a big pile of pdf documents on us and asked us to extract information from them. I thought I would give Mat2Vec a try, since it's interested me for the last 6 months.