metadata

tags:
  - exbert
language:
  - hr
pipeline_tag: text-classification
widget:
  - text: >-
      Uzbuđen sam što mogu podijeliti da smo postigli novi rekord u prodaji
      naših proizvoda ovog kvartala.
  - text: >-
      Na web stranici je objavljena nova analiza tržišta koja pokazuje nove
      trendove u sektoru nekretnina.
  - text: >-
      S velikim razočaranjem moram reći da su korisničke recenzije našeg
      najnovijeg proizvoda izrazito loše.
datasets:
  - dejanseo/croatian-sentiment

Croatian Sentiment

Croatian sentence sentiment classification model. A fine-tuned version of XLM-RoBERTa Large, trained and released by Dejan Marketing.

Sentiment Labels

0: positive
1: neutral
2: negative

Engage Our Team

Interested in using this in an automated pipeline for bulk link prediction?

Please book an appointment to discuss your needs.

Training Data:

Synthetic dataset generated using mistralai/Mistral-7B-Instruct-v0.3