LLM ChatBot with Gradio UI Tutorial - Part 2

In this Applied NLP LLM Tutorial, We will build our Custom KnowledgeBot using LLama-Index and LangChain. LangChain for accessing Hugging Face Model Hub and GPT-Index for Vector Index Building and Index Querying - Information Retrieval. Gradio for ChatBOT User Interface

❤️ If you want to support what we are doing ❤️
Support here:

[NadaaTaiyab]

Thank you for showing your mistakes and fixes in this video…. And I hope your head is intact after that 😂 But seriously you are very good at explaining this stuff and your manner is engaging so I enjoy watching your videos. Thank you!

[projectbit]

Damn! Your like a creator that can read what's going in our minds! :D

[NillsBoher]

Thanks! I have a question: How do I make the gradio chatbot to remember the history and use it as a context? For example if a user asked to answer in french, I want the chat to remember it for all the next quesrios. Right now I see every question is atomic - it dosnt remeber that it should answer in French...

[bhushankapkar]

Thanks for creating great content, few queries 1. In case we wanted to give feel of small window of chatbot using gradio or it will always take complete space of web browser, is it possible with gradio. 2. In case we want to make use of the function and want to make production ready chatbot which framework is recommended can you take that as reference in upcoming videos. If possibility 3. In GPT Index if documents is too long as soon as you increase size of more that 2048 vectors it's doesn't allow to get it trained and gives error.

[geekyprogrammer]

your consistency is amazing brother! May Allah bless you!

[vmartme]

Thank you so much friend! Greetings from EU :)

[conanssam]

Thanks, I'm trying to create a special chatbot with unique information in GPT.
I don't know if I understood well, but if I put openai api key and huggingface api key in colab shared notebook, it should work?

[sureshchittewar]

Great. Can we similarly do it with Vicuna or Stable LM model ?

[yonbek]

Great content. It would help a lot if you post a step bt step instructions in pdf on description page.

[CristianQuitian-nu]

Hi! thanks for the video.
I am trying to build a gradio chatbot with custom data and streaming based on your code.
My database is really big and it takes usually 40 seconds for chatgpt to reply.
So I added your streaming code.
It works but the behaviour is not as I expected.
It processes the answer for 40 seconds and then, the text appears like it is being streamed.
is this really how it should behave? shouldn't it load the characters as the answer is being generated?

[tusharkhatri]

Here you are saving full chat history but after some time api token 4096 will reach so in m case i used conversational buffer window memory how to use that in gradio?

[haticeobuz]

Hello again, the index used in the tutorial is no longer working since GPTIndex is now LlamaIndex. Are you planning on making a video of the new version?

[robertoamarillas]

Amazing, but, how to put this on a website?

[haticeobuz]

Hello, I am planning to create a bot to extract some information from a given text. The flan-t5-large model did not perform the best for this task. Do you have any model recommendations for this specific task?

[lucifergaming]

Bro can you help me .
There is 3d image inpainting github in which they provide official colab notebook but they never updated it. The worked up until yesterday but it stopped working today. I think colab has updated some settings so now notebook does not work. I am damn sure that issue is just minor but i cannot fix it cuz i dont know coding. So please help.

[conanssam]

ValueError: Error raised by inference API: Authorization header is correct, but the token seems invalid

I'm getting this error, what should I fix when the token is invalid?

[amiteshpuri]

can you integrate gradio chatbot option in this demo?

[evilcorp]

You realise that you dont link to part one right?