chiwacych/hate_speech_identifier-Kenya

Hate Speech Identifier — Kenya

A fine-tuned BERT classifier that labels input text as Hate/Offensive Speech or Non-Hate Speech. The positive class follows the Class column of the training data, which is driven largely by offensive_language rather than hate_speech alone - so it flags offensive content more broadly than hate speech strictly.

The model weights live in Hate_speech_model/ in this repo and are loaded locally by app.py.

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference