From the course: Fundamentals of AI Engineering: Principles and Practical Applications
Join today to access over 24,300 courses taught by industry experts.
Enhancing retrieval with reranking
From the course: Fundamentals of AI Engineering: Principles and Practical Applications
Enhancing retrieval with reranking
- [Instructor] In our previous videos, we explored how hybrid retrieval combines different retrieval methods like BM25 and vector search. Today, we're going to introduce the concept of re-ranking. To get started, open chapter_6, and open up the notebook titled 06_04.ipynb. As always, in the upper right-hand side of your notebook, make sure you've selected the .venv virtual environment. Today, we're going to take our retrieval pipeline to the next level by introducing re-ranking. Re-ranking is a powerful technique that can significantly improve the quality of your search results by applying a more sophisticated model to refine the initial set of retrieved documents. Let's get started with a fundamental concept called re-ranking. Fundamentally, re-ranking is a two stage process, the initial retrieval where we use efficient methods to get a candidate set of potentially relevant documents, and then the re-ranking step where we apply a more sophisticated model to reorder these candidates…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.