castorini/afribertalarge fine-tuned on the Kiswahili subset of AfriSenti (Muhammad et al., 2023) to classify tweets as positive / neutral / negative. best epoch selected on validation macro-F1 (seed 42). Demo: https://huggingface.co/spaces/RubaxTyra/kiswahili-sentiment-demo Limitations: trained on only 1,810 tweets; emojis and punctuation were removed from the training data; struggles with sarcasm, slang/Sheng and with negative news vs negative opinion.
Independent publisher
Rubax Tyra
RubaxTyra
Models in Library1
Datasets in Library0
Models on Hugging Face1
Followers—
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Text classification.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.