Semantic Reranking with Cohere Reranker

This example will show how to combine search and semantic reranking to improve the accuracy of your search results. We'll be using the rerank feature from Cohere.

Note: for a complete integration with Cohere please refer to this notebook. This example focuses on Cohere reranking only through an Elastic retriever query.

Requirements

For this example, you will need:

• An Elastic deployment:

  • We'll be using Elastic Cloud for this example (available with a free trial)

• Elasticsearch 8.15 or above, or Elasticsearch serverless

• Cohere API key

Create Elastic Cloud deployment

If you don't have an Elastic Cloud deployment, sign up here for a free trial.

Install packages and connect with Elasticsearch Client

To get started, we'll need to connect to our Elastic deployment using the Python client (version 8.15.0 or above). Because we're using an Elastic Cloud deployment, we'll use the Cloud ID to identify our deployment.

First we need to pip install the elasticsearch package:

Next, we need to import the modules we need.

🔐 NOTE: getpass enables us to securely prompt the user for credentials without echoing them to the terminal, or storing it in memory.

Now we can instantiate the Python Elasticsearch client.

First we prompt the user for their password and Cloud ID. Then we create a client object that instantiates an instance of the Elasticsearch class.

Enable telemetry

Knowing that you are using this notebook helps us decide where to invest our efforts to improve our products. We would like to ask you that you run the following code to let us gather anonymous usage statistics. See telemetry.py for details. Thank you!

Test the Client

Before you continue, confirm that the client has connected with this test.

Refer to the documentation to learn how to connect to a self-managed deployment.

Read this page to learn how to connect using API keys.

Set up Cohere Inference Endpoint

We'll be using the Cohere rerank feature to perform semantic reordering of search hits through an Elasticsearch inference endpoint.

Go to the Cohere website and create an API key, then set it here.

Create the Inference Endpoint

Let's create the inference endpoint by using the Create inference API.

For this example we'll use the Cohere service, but the inference API also supports many other inference services.

Create the Index

Now we need to create an index. Let's create one that enables us to perform search and semantic reranking on text articles.

Populate the Index

Let's populate the index with a couple of random article fragments from Wikipedia.

Search without reranking

First let's run a classic search that uses lexical text matching.

Aside: Pretty printing Elasticsearch search results

Your search API calls will return hard-to-read nested JSON. We'll create a little function called pretty_search_response to return nice, human-readable outputs from our examples.

Assume we're interested to learn about the solar eclipse, but we don't know the exact name of this phenomenon. We'll perform a classic search that matches the text "the Moon covers the Sun". Let's see what results this finds:

The top hits - Cheshire, Sun Moon Lake, Unification Church and so on - all come up because the text has matching words with our query's words, for example "covers" or "sun". However, these contents are unrelated to the meaning of our query. Further down below, result #7 is an article about the solar eclipse, but it's lost among the many other hits.

Can we somehow get more relevant results?

Search with reranking

Enter semantic reranking! We'll instruct Elasticsearch to run the same query, but this time also perform semantic reranking on the top results. For this we need to wrap our query in a text_similarity_reranker retriever, and reference the previously created Cohere inference endpoint that will do the reranking.

Much better! Not only are the top results semantically close to our query "the Moon covers the Sun", the irrelevant results with a low score were discarded from the response. As a result, the list of articles we ended up with are indeed those that provide the best answer to our question.

What's also great about reranking is that it can be used on top of existing search solutions out of the box. Under the hood the same lexical search was executed as before - the one that resulted in mixed hits -, then Cohere took the texts from the top articles and reordered them according to their relation to our query's meaning.

Whether your search application uses lexical, vector or hybrid search, reranking can improve your results.

Conclusion

Semantic reranking is an incredibly powerful tool for boosting the performance of a search experience or a RAG tool. It lets us immediately add semantic search capabilities to existing Elasticsearch installations out there.