Ever felt lost when a sommelier throws out words like "Rhubarb hangs out with cherry cider and the mellowest cinnamon" to describe a wine? Well - me too!
At the end of last week I was on a quest to bridge the gap between flowery descriptions and actual grape varieties! Inspired by the book "NLP and Transformers", I decided to fine-tuned a DistilBERT encoder model to do just that. I was curious just how good a relatively small model could be on raw data with minimal data wrangling.
The results surprised me. It worked better than expected! If you're interested in seeing the code and how good the model is, check out the [Jupyter notebook], datasets and models.