Show HN: Autofit2 – End-to-end pipeline for multilingual text classification
Summary
Autofit2 is a public, open-source project delivering an end-to-end pipeline for multilingual text classification built on setfit and SBERT embeddings. It supports few-shot learning with high accuracy, multilingual models, and a JSON-config-driven workflow that handles data loading, training, evaluation, and deployment, with artifacts like a deployable model archive and a model card.