Open in Colab

Reranking is a technique that leverages embeddings as the last stage of a retrieval process, and is especially useful in RAG systems.

We can rerank results from semantic search as well as any other search systems such as lexical search. This means that companies can retain an existing keyword-based (also called “lexical”) or semantic search system for the first-stage retrieval and integrate the Rerank endpoint in the second-stage reranking.

In this tutorial, you’ll learn about:

  • Reranking lexical/semantic search results
  • Reranking semi-structured data
  • Reranking tabular data
  • Multilingual reranking

You’ll learn these by building an onboarding assistant for new hires.

Setup

To get started, first we need to install the cohere library and create a Cohere client.

PYTHON
1# pip install cohere numpy
2
3import numpy as np
4import cohere
5
6co = cohere.Client("COHERE_API_KEY") # Get your API key: https://dashboard.cohere.com/api-keys

Reranking lexical/semantic search results

Rerank requires just a single line of code to implement.

Suppose we have a list of search results of an FAQ list, which can come from semantic, lexical, or any other types of search systems. But this list may not be optimally ranked for relevance to the user query.

This is where Rerank can help. We call the endpoint using co.rerank() and pass the following arguments:

  • query: The user query
  • documents: The list of documents
  • top_n: The top reranked documents to select
  • model: We choose Rerank English 3
PYTHON
1# Define the documents
2faqs_short = [
3 {"text": "Reimbursing Travel Expenses: Easily manage your travel expenses by submitting them through our finance tool. Approvals are prompt and straightforward."},
4 {"text": "Working from Abroad: Working remotely from another country is possible. Simply coordinate with your manager and ensure your availability during core hours."},
5 {"text": "Health and Wellness Benefits: We care about your well-being and offer gym memberships, on-site yoga classes, and comprehensive health insurance."},
6 {"text": "Performance Reviews Frequency: We conduct informal check-ins every quarter and formal performance reviews twice a year."}
7]
PYTHON
1# Add the user query
2query = "Are there fitness-related perks?"
3
4# Rerank the documents
5results = co.rerank(query=query,
6 documents=faqs_short,
7 top_n=2,
8 model='rerank-english-v3.0')
9
10print(results)
id='9633b278-93ff-4664-a142-7d9dcf0ec0e5' results=[RerankResponseResultsItem(document=None, index=2, relevance_score=0.01798621), RerankResponseResultsItem(document=None, index=3, relevance_score=8.463939e-06)] meta=ApiMeta(api_version=ApiMetaApiVersion(version='1', is_deprecated=None, is_experimental=None), billed_units=ApiMetaBilledUnits(input_tokens=None, output_tokens=None, search_units=1, classifications=None), tokens=None, warnings=None)
PYTHON
1# Display the reranking results
2def return_results(results, documents):
3 for idx, result in enumerate(results.results):
4 print(f"Rank: {idx+1}")
5 print(f"Score: {result.relevance_score}")
6 print(f"Document: {documents[result.index]}\n")
7
8return_results(results, faqs_short)
Rank: 1
Score: 0.01798621
Document: {'text': 'Health and Wellness Benefits: We care about your well-being and offer gym memberships, on-site yoga classes, and comprehensive health insurance.'}
Rank: 2
Score: 8.463939e-06
Document: {'text': 'Performance Reviews Frequency: We conduct informal check-ins every quarter and formal performance reviews twice a year.'}

Further reading:

Reranking semi-structured data

The Rerank 3 model supports multi-aspect and semi-structured data like emails, invoices, JSON documents, code, and tables. By setting the rank fields, you can select which fields the model should consider for reranking.

In the following example, we’ll use an email data example. It is a semi-stuctured data that contains a number of fields – from, to, date, subject, and text.

Suppose the new hire now wants to search for any emails about check-in sessions. Let’s pretend we have a list of five emails retrieved from the email provider’s API.

To perform reranking over semi-structured data, we add an additional parameter, rank_fields, which contains the list of available fields.

The model will rerank based on order of the fields passed in. For example, given rank_fields=['title','author','text'], the model will rerank using the values in title, author, and text sequentially.

PYTHON
1# Define the documents
2emails = [
3 {"from": "hr@co1t.com", "to": "david@co1t.com", "date": "2024-06-24", "subject": "A Warm Welcome to Co1t!", "text": "We are delighted to welcome you to the team! As you embark on your journey with us, you'll find attached an agenda to guide you through your first week."},
4 {"from": "it@co1t.com", "to": "david@co1t.com", "date": "2024-06-24", "subject": "Setting Up Your IT Needs", "text": "Greetings! To ensure a seamless start, please refer to the attached comprehensive guide, which will assist you in setting up all your work accounts."},
5 {"from": "john@co1t.com", "to": "david@co1t.com", "date": "2024-06-24", "subject": "First Week Check-In", "text": "Hello! I hope you're settling in well. Let's connect briefly tomorrow to discuss how your first week has been going. Also, make sure to join us for a welcoming lunch this Thursday at noon—it's a great opportunity to get to know your colleagues!"}
6]
PYTHON
1# Add the user query
2query = "Any email about check ins?"
3
4# Rerank the documents
5results = co.rerank(query=query,
6 documents=emails,
7 top_n=2,
8 model='rerank-english-v3.0',
9 rank_fields=["from", "to", "date", "subject", "body"])
10
11return_results(results, emails)
Rank: 1
Score: 0.1979091
Document: {'from': 'john@co1t.com', 'to': 'david@co1t.com', 'date': '2024-06-24', 'subject': 'First Week Check-In', 'text': "Hello! I hope you're settling in well. Let's connect briefly tomorrow to discuss how your first week has been going. Also, make sure to join us for a welcoming lunch this Thursday at noon—it's a great opportunity to get to know your colleagues!"}
Rank: 2
Score: 9.535461e-05
Document: {'from': 'hr@co1t.com', 'to': 'david@co1t.com', 'date': '2024-06-24', 'subject': 'A Warm Welcome to Co1t!', 'text': "We are delighted to welcome you to the team! As you embark on your journey with us, you'll find attached an agenda to guide you through your first week."}

Reranking tabular data

Many enterprises rely on tabular data, such as relational databases, CSVs, and Excel. To perform reranking, you can transform a dataframe into a list of JSON records and use Rerank 3’s JSON capabilities to rank them.

Here’s an example of reranking a CSV file that contains employee information.

PYTHON
1import pandas as pd
2from io import StringIO
3
4# Create a demo CSV file
5data = """name,role,join_date,email,status
6Rebecca Lee,Senior Software Engineer,2024-07-01,rebecca@co1t.com,Full-time
7Emma Williams,Product Designer,2024-06-15,emma@co1t.com,Full-time
8Michael Jones,Marketing Manager,2024-05-20,michael@co1t.com,Full-time
9Amelia Thompson,Sales Representative,2024-05-20,amelia@co1t.com,Part-time
10Ethan Davis,Product Designer,2024-05-25,ethan@co1t.com,Contractor"""
11data_csv = StringIO(data)
12
13# Load the CSV file
14df = pd.read_csv(data_csv)
15df.head(1)

Here’s what the table looks like:

namerolejoin_dateemailstatus
Rebecca LeeSenior Software Engineer2024-07-01rebecca@co1t.comFull-time

Below, we’ll get results from the Rerank endpoint:

PYTHON
1# Define the documents and rank fields
2employees = df.to_dict('records')
3rank_fields = df.columns.tolist()
4
5# Add the user query
6query = "Any full-time product designers who joined recently?"
7
8# Rerank the documents
9results = co.rerank(query=query,
10 documents=employees,
11 top_n=1,
12 model='rerank-english-v3.0',
13 rank_fields=rank_fields)
14
15return_results(results, employees)
Rank: 1
Score: 0.986828
Document: {'name': 'Emma Williams', 'role': 'Product Designer', 'join_date': '2024-06-15', 'email': 'emma@co1t.com', 'status': 'Full-time'}

Multilingual reranking

The Rerank endpoint also supports multilingual semantic search via the rerank-multilingual-... models. This means you can perform semantic search on texts in different languages.

In the example below, we repeat the steps of performing reranking with one difference – changing the model type to a multilingual one. Here, we use the rerank-multilingual-v3.0 model. Here, we are reranking the FAQ list using an Arabic query.

PYTHON
1# Define the query
2query = "هل هناك مزايا تتعلق باللياقة البدنية؟" # Are there fitness benefits?
3
4# Rerank the documents
5results = co.rerank(query=query,
6 documents=faqs_short,
7 top_n=2,
8 model='rerank-multilingual-v3.0')
9
10return_results(results, faqs_short)
Rank: 1
Score: 0.42232594
Document: {'text': 'Health and Wellness Benefits: We care about your well-being and offer gym memberships, on-site yoga classes, and comprehensive health insurance.'}
Rank: 2
Score: 0.00025118678
Document: {'text': 'Performance Reviews Frequency: We conduct informal check-ins every quarter and formal performance reviews twice a year.'}

Conclusion

In this tutorial, you learned about:

  • How to rerank lexical/semantic search results
  • How to rerank semi-structured data
  • How to rerank tabular data
  • How to perform Multilingual reranking

We have now seen two critical components of a powerful search system - semantic search, or dense retrieval (Part 4) and reranking (Part 5). These building blocks are essential for implementing RAG solutions.

In Part 6, you will learn how to implement RAG.