This folder contains notebooks that demonstrate how to work with Hugging Face hosted models in Elasticsearch.
The following notebooks are available:
In the loading-model-from-hugging-face.ipynb
notebook you'll learn how to:
- Use the
eland
tool to import theall-MiniLM-L6-v2
model from Hugging Face. - Index documents into Elasticsearch and generate text embeddings using an ingest pipeline.
- Transform search queries into vectors using the Hugging Face model and perform vector searches over the dataset.