README.md

---
library_name: transformers
pipeline_tag: text-classification
tags:
- biology
- herbarium
- location
language:
- en
---
![RBG Kew Logo](https://c.ststat.net/Content/Sites/kew/generic/images/logo.png)
![RBG Kew Herbarium Packets](https://www.kew.org/sites/default/files/styles/read_watch_listing/public/2019-02/herbarium%20specimens.png.webp?itok=XByp1zeV)

RoBERTa for binary sequence classification fine-tuned to classify text derived from herbarium packets as location sensitive.
Fine-tuned with 500,000 cleaned data samples from RBG Kew's Herbarium dataset available on GBIF (https://doi.org/10.15468/ly60bx).
Trained primarily for English language but may work with other languages due to the large variety of text present in the Kew Herbarium.