This project demonstrates the use of Elasticsearch for vector search tasks. The provided Jupyter Notebook, vector_search.ipynb, loads an embeddings model using sentence-transformers, creates embeddings for the combined question and answer field, and creates a Vector Search Engine to search for user questions in FAQ documents. We evaluate the search engine results using hit rate and mean reciprocal rank (MRR). Finally, evaluate Elasticsearch, which is more efficient as it uses approximate techniques rather than computing similarity with all the vectors, producing identical results.
To set up the project, follow these steps:
pip install pipenv
pipenv installThen, select the pipenv environment in the Jupyter kernel.
To run the Elasticsearch Docker container, use the following command:
docker run -it \
--rm \
--name elasticsearch \
-p 9200:9200 \
-p 9300:9300 \
-e "discovery.type=single-node" \
-e "xpack.security.enabled=false" \
docker.elastic.co/elasticsearch/elasticsearch:8.4.3