interactive-rag
ФоркИзбранное
0
0
Описание
Using MongoDB Atlas + ActionWeaver, we can build a user proxy agent that efficiently retrieves and ingests relevant information. This agent presents the data to the user in an interactive and customizable manner, enhancing the overall user experience.
Языки
- Python100%
Fabian Valle
19 фев 2024, 09:18
3 года назад
2 года назад
3 года назад
3 года назад
2 года назад
3 года назад
3 года назад
3 года назад
3 года назад
3 года назад
INTRODUCTION
Updates
Chunking text: Summarize or store the raw data?
VIDEO
FLOW
Getting Started
Usage
Interactive Retrieval Augmented Generation
The "Ingest Process"
BUILDING A RAG-AGENT
Why Choose ActionWeaver?
ActionWeaver Basics: What is an Agent anyway?
Key features of ActionWeaver include:
Key features of OpenAI functions include:
ActionWeaver Basics: actions
ActionWeaver Basics: stop=True
ReAct Prompt Technique + Chain of Thought
DEMO
Tell the bot which results to learn from:
Change RAG strategy
List All Sources
Remove a source of information
Credit
Additional MongoDB Resources
Additional Reading
Contributing
License
README.md