huggingface/setfit
SetFit is a few-shot fine-tuning framework for Sentence Transformers that achieves high accuracy with minimal labeled data.
Velocity · 7d
+1.9
★ / day
Trend
→steady
star history
SetFit enables efficient few-shot learning by fine-tuning sentence transformer models without requiring handcrafted prompts or verbalizers. It generates rich embeddings from text examples and supports multilingual classification. The framework is significantly faster than large-scale few-shot methods like T0 or GPT-3 while remaining competitive in accuracy.