Skip to content

gulsensabak/LLM-ReAct-Framework

Repository files navigation

ReAct

ReAct: Synergizing Reasoning and Acting in Language Models

What is ReAct?

It is all about reasoning and action

It advanced the way of thinking from standard sort of prompting through chain of thought prompting actually now being able to have reason tracing and actions as well.

We are getting multi-step ways of thinking but not just in one shot (like chain of thought) but over multiple shots with this.

Code Files

  • chainOfThoughtReasoning.ipynb: Explains how chain of thought reasoning works in large language models.

  • manualReact.pynb: The way the ReAct works is that using tools If we don't give any tools to the ReAct what happens? I wrote my own Manual ReAct in that code file to see what happens if I don't give any tool to React.

  • react.ipynb: I used React with Langchain. I put multi-hop example to see how react works.

  • customizeTools.ipynb: We can specialize the Tools in Agents that comes from Langchain library. I added one more tool to see how it works.

  • customizeTools2.ipynb: I created my own tool to solve algebraic operations on complex numbers. Also, I used that tool with Langchain Agent.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published