Skip to content

llmpro70b/Chat-Visualizer

Repository files navigation

Chat Visualizer of AI Agent Multi-Turn Dialogue

In this tutorial, I will introduce how to use the chat visualizer feature of agentboard (Github,Docs) to help visualize huge amount of chat logs. agentboard is a python package to help visualize agent_loop, RAG, Chat messages, and multi-modal datas. It provides useful web GUI to quickly protytype the Chat logs, and you can take screen shots of the demos, put them in your tech report, arxiv papers, or even continue the Chat of some open ended funny dialogues.

Example 1: Simple Demo of using the AgentBoard to log the messages.

ab.summary.messages

A typical chat history between AI agents and users are in the format of a list of jsons. [{"role": "user", "content": "message 1"}, {"role": "assistant", "content": "message 2"}, ...]


import agentboard as ab

def fun_chat_history():
    """
        Chat logs between a user and a chatbot
    """
    chat_history = []
    chat_history.append({"role": "user", "content": "Please tell me a joke"})
    chat_history.append({"role": "assistant", "content": "Why don’t skeletons fight each other? Because they don’t have the guts! "})
    chat_history.append({"role": "user", "content": "It's not funny. Try another one."})
    chat_history.append({"role": "assistant", "content": "Why do programmers prefer dark mode? Because light attracts bugs! "})
    chat_history.append({"role": "user", "content": "Alright, this one is good..."})
    return chat_history

def run_chat_visualizer():
    """
        agentboard --logdir=./log
        # agentboard --logdir=./log --logfile=xxx.log --static=./static --port=5000
    """
    messages = fun_chat_history()
    with ab.summary.FileWriter(logdir="./log", static="./static") as writer:
        ab.summary.messages(name="Fun Chat Log", data=messages, agent_name="ChatGPT")

if __name__ == "__main__":
    run_chat_visualizer()


Run the Demo

    python run_chat_visualizer.py

You will find the logs will be saved to local "log" folder

Then you can run the agentboard and see the visualized chat history in a Chatbot.

agentboard --logdir=./log

Chat Visualizer Using Deepnlp Agentboard

Alternatively, you can also use the online Chat Visualizer Tool for quick prototypes (DeepNLP Chat Visualizer) It provides more UI choices, such as the ones like ChatGPT, Whatsapp, Wechat, etc.

Agents Related Pipeline Workflow and Document

AI Services Reviews and Ratings

AI Agent

AI Agent User Reviews
Microsoft AI Agents Reviews
Claude AI Agents Reviews
OpenAI AI Agents Reviews
AgentGPT AI Agents Reviews
Saleforce AI Agents Reviews
Google AI Agents Reviews
Google AI Agents Space

Chatbot

OpenAI o1 Reviews
ChatGPT User Reviews
Gemini User Reviews
Perplexity User Reviews
Claude User Reviews
Qwen AI Reviews
Doubao Reviews
ChatGPT Strawberry
Zhipu AI Reviews

AI Image Generation

Midjourney User Reviews
Stable Diffusion User Reviews
Runway User Reviews
GPT-5 Forecast
Flux AI Reviews
Canva User Reviews

AI Video Generation

Luma AI
Pika AI Reviews
Runway AI Reviews
Kling AI Reviews
Dreamina AI Reviews

AI Education

Coursera Reviews
Udacity Reviews
Grammarly Reviews

Robotics

Tesla Cybercab Robotaxi
Tesla Optimus
Figure AI
Unitree Robotics Reviews
Waymo User Reviews
ANYbotics Reviews
Boston Dynamics

AI Tools

DeepNLP AI Tools

AI Widgets

Apple Glasses
Meta Glasses
Apple AR VR Headset
Google Glass
Meta VR Headset
Google AR VR Headsets

Social

Character AI

Self-Driving

BYD Seal
Tesla Model 3
BMW i4
Baidu Apollo Reviews
Hyundai IONIQ 6

Related Blogs

AgentBoard AI Agent Visualization Toolkit
DeepNLP AI Agents Designing Guidelines
Introduction to multimodal generative models
Generative AI Search Engine Optimization
AI Image Generator User Reviews
AI Video Generator User Reviews
AI Chatbot & Assistant Reviews
Best AI Tools User Reviews
AI Boyfriend User Reviews
AI Girlfriend User Reviews

About

AI Agent Chat Visualizer Visualize the Chat Data

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published