utilities for loading and running text embeddings with onnx

Sentence Transformers now appears to have robust support for running arbitrary embedding models with ONNX backend. As such, I would recommend using Sentence Transformers for embeddings inference with ONNX. This package has not been maintained and has some bugs affecting accuracy.

0.0.14 - 2024-06-19

• support for loading from any huggingface repo with from_pretrained (provided it has a tokenizer and some onnx model in it)
• added mxbai-embed-large to the model registry
• embedding model .encode method now returns numpy arrays by default instead of lists