The FASTEST way to build CHAT UI for LLAMA-v2
In this video, ill show you the fastest way of building a chatbot user interface (chat ui)! We will be using gradio.
Please subscribe and like the video to help me keep motivated to make awesome videos like this one. :)
My book, Approaching (Almost) Any Machine Learning problem, is available for free here: bit.ly/approachingml
Follow me on:
Twitter: / abhi1thakur
LinkedIn: / abhi1thakur
Kaggle: kaggle.com/abhishek
Пікірлер: 50
Code is available here: gist.github.com/abhishekkrthakur/a3d712c2709dcfbfa6a803fcfbcb5737 Please subscribe to help me keep motivated to make awesome videos like this one. :)
Thank you so much for coming back to your KZread channel. Now we can learn a lot🎉
I am working in data science since last year. I am regularly following ur videos. I feel that u have GREAT content but the level that you teach you is much higher than the aveage ppl(according to me, could b wrong) so if u could plz teach us in a beginner way then it would be very much helpful thank you again!!!! love from india bhai!
Keep good work Abhishek 👍👍👍 Great videos on LLMs.
the code seems to work but as i open the gradio url to test it, its showing error as soon as i put in some question
Thanks! Great work! Could you tell me about set-up on VScode? Docker?
Amazing content. Thanks
Never-minding for a sec that I really appreciate the content of this video, I would love to know how you're doing your recording... especially how are you getting your talking head cutout to superimpose on the video.
i am using python and having llama-2-7b-chat on my linux machine. how can i load the model in above UI. waiting for a response.
Hi Abhi, could you pls drop a link of video where we can deploy model by ourselves? Downloaded Llama 2 model to Any Cloud end point?
Hi Abishek, Thank you for the great content. I just tried your code but I got the same error while prompting . tell me pls Thanks
Is there a chance to make a video to query a pdf or chat with data will llama2
Can we build a user interface to let llama2 analyze data in excel spreadsheets?
Hi Abhishek, great video! Quick question, are there any CPU requirements to run this locally?
@abhishekkrthakur
Жыл бұрын
not really. but note that the endpoint will require GPU and for that, you can take a look at this tutorial: kzread.info/dash/bejne/oWV2pdNqe627fLA.html
Great video, Abhishek. Is it possible to host ChatUI on Azure Webapp service? Have you come a cross something like that? Thanks! 🙏
hey friend. it seems that you know a lot!! could you give me a hand with this? I want to know what's the best way of creating your own "chatgpt" but giving a specific amount of X files. Let's say a small book, and then be able to ask questions about it, but running it locally or in the cloud (not using any third party basically). what's the best way of doing it? is it even possible? like running something like that in google colab with files from google drive or something like that? thanks in advance man!
👍👍👍
This is great to build a quick poc in 2 hours for presentation
@abhishekkrthakur
Жыл бұрын
takes 20mins or less. hehe
@anandteerthrparvatikar5359
Жыл бұрын
Absolutely, even ppt content from llama... However, Company specific template :)
Can we build a private chatbot using llama 2? The way you developed in your last video using Falcon 40B?
@abhishekkrthakur
Жыл бұрын
all you need to do is replace the model and use the instructions shown in this video :)
Great Content! One question what's LLAMA_70B linked to? I have a llama-2-7b folder downloaded into my local dev. How can I link that into the os.environ()? Thanks
@abhishekkrthakur
Жыл бұрын
thanks! described here: kzread.info/dash/bejne/oWV2pdNqe627fLA.html
@user-qx7jx1qr4g
Жыл бұрын
@@abhishekkrthakur how exactly does this work? It would be super helpful to know. Great work overall!
I am sorry to ask, But I didnt understand what you did at 15:35 to resolve that error? you said that you are exporting the environment variables but why and where?
@abhishekkrthakur
Жыл бұрын
in the terminal i wrote: export LLAMA_70B=........ where ........ is the path to your endpoint. in case you want to learn how to create this endpoint, check out the other videos in my LLM playlist: kzread.info/head/PL98nY_tJQXZlXLELjCMA8cciKLRE2eLME
@koshalsingh9786
Жыл бұрын
@@abhishekkrthakur thanks for such quick reply
Can your code handle multiple requests concurrently ? (multiple users asking a question at the same time), is this flask app ?
@abhishekkrthakur
Жыл бұрын
yes. and no, its not a flask app
@sj0998
Жыл бұрын
@@abhishekkrthakurthanks 😊
Thank you for this tutorial. Please, is a video turial on how to sharred a model. I want to sharred this model togethercomputer/Llama-2-7B-32K-Instruct
Are you planning on making a chatbot for Llama 3?
I laughed at that one line statement of gradio to create a chat interface. Need to try this. Does this format LaTeX? Also, how is 70B running so fast on your machine?
@abhishekkrthakur
Жыл бұрын
its not running on my machine 🙂 im not surr about latex, unfortunately. worth a try
@foreignconta
Жыл бұрын
@@abhishekkrthakur I see. Will try!
@aekanshgupta6642
11 ай бұрын
@@abhishekkrthakur Where is it running? If not on your machine. Please let me know, very crucial.
Now let's finetune it and do some RAG
@ChefDomein
Жыл бұрын
Yes this please!🙏
@mitalicops8538
2 ай бұрын
Hi there what do mean by fine tuning and RAG plz can u explain it, i am a beginner.
@denzilstudios7072
2 ай бұрын
@@mitalicops8538 Retrieval Augmented Generation (RAG) first searches through a bunch of documents using queries, which are basically questions or keywords. It then organizes all that info into a special kind of database called a vector database. When you ask a question using RAG, it quickly finds the best answer from that database and helps you generate new text based on it.
This keeps giving an error message . When this happened to you, you said That you forgot to export the environment variable WHICH IS????
Are it support Arabic language?
You forgot to describe principal part of the item - server side ...
It will be useful to have link for github repo
@abhishekkrthakur
Жыл бұрын
here you go: gist.github.com/abhishekkrthakur/a3d712c2709dcfbfa6a803fcfbcb5737
Thank you for the great work!! But I'm facing the following error: requests.exceptions.ConnectionError: HTTPConnectionPool(host='localhost', port=3000): Max retries exceeded with url: / (Caused by NewConnectionError(': Failed to establish a new connection: [WinError 10061] Aucune connexion n’a pu être établi e car l’ordinateur cible l’a expressément refusée')) can you help please?
@koshalsingh9786
Жыл бұрын
did that error got resolved?
@youssefmellah2525
Жыл бұрын
@@koshalsingh9786 no not yet bro