Sentence Transformers

概要

Sentence Transformers provides an easy method to compute embeddings for accessing, using, and training state-of-the-art embedding and reranker models. It allows users to generate embeddings using Sentence Transformer models or to calculate similarity scores using Cross-Encoder models, unlocking applications like semantic search, semantic textual similarity, and paraphrase mining. With a wide selection of pre-trained models and the ability to train custom models, Sentence Transformers offers a versatile solution for various natural language understanding tasks.

主な機能

• Compute embeddings using Sentence Transformer models
• 16,712 GitHub stars
• Supports various transformer networks including BERT, RoBERTa, XLM-R, DistilBERT, Electra, BART
• Access over 10,000 pre-trained models on Hugging Face
• Calculate similarity scores using Cross-Encoder (reranker) models
• Train or finetune custom embedding and reranker models

ユースケース

• Semantic Search
• Semantic Textual Similarity
• Paraphrase Mining