mirror of
https://github.com/qdrant/landing_page.git
synced 2026-09-25 22:18:30 +02:00
nastyapash
09877ed06e
* Add RAG Landing Page * fix * fix form, image and links * fix * Content update * shortened file link * update links
42 lines
1.2 KiB
Markdown
42 lines
1.2 KiB
Markdown
---
|
|
title: How to evaluate a RAG system
|
|
text: This guide will teach you how to evaluate a RAG system for both accuracy and quality.
|
|
subtitle: Stages prone to errors
|
|
listTitle: "You will learn to maintain RAG performance by testing for:"
|
|
list:
|
|
- id: 0
|
|
text: Search precision
|
|
- id: 1
|
|
text: Recall
|
|
- id: 2
|
|
text: Contextual relevance
|
|
- id: 3
|
|
text: Response accuracy.
|
|
image:
|
|
src: /img/rag-evaluation-guide/stages.svg
|
|
alt: Stages
|
|
features:
|
|
- id: 0
|
|
icon:
|
|
src: /icons/outline/search-text-blue.svg
|
|
alt: Search text
|
|
title: Information retrieval
|
|
description: This stage involves searching and fetching relevant information from a knowledge base or external sources.
|
|
- id: 1
|
|
icon:
|
|
src: /icons/outline/integration-blue.svg
|
|
alt: Integration
|
|
title: Information augmentation
|
|
description: In this stage, the retrieved information is processed and combined with the original query
|
|
- id: 2
|
|
icon:
|
|
src: /icons/outline/vectors-blue.svg
|
|
alt: Vectors
|
|
title: Generating responses
|
|
description: Using the augmented information, the language model generates a response to the original query.
|
|
downloadGuide:
|
|
text: Download the Guide
|
|
url: "#form"
|
|
sitemapExclude: true
|
|
---
|