Very easy and perfect way to explain, thanks mam. Please come up with fine tuning, deployment on cloud, how to test llm model's performance kind of videos. Your way of explanation is very simple and effective.
Thank you for a detailed and easy-to-understand tutorial. I have a request, please create a tutorial to implement RAG on any LLM and on any document, such as text files and databases. I'll surely do research from my side, but a help from mentor speeds up the learning process.
I followed your steps in installation of pytorch and torch vision and open cv yolo v5 etc..success fully I installed opencv2 with cuda support and installed successfully pytorch and torch vision etc but in installation of yolo v5 I got struck at ultralytics then I got know python 3.6 is not compatible with installation of ultralytics. Now I again flashed os into board and I created new virtual environment with python 3.8 now pytorch is not getting installed as my jetpack SDK is 4.5 and I cannot upgrade to 5 as hardware is not supported, but in many articles it is said that ultralytics is important how do I handle this situation 😅
thank you for the video . I like your consistency in moving with GenAI and LLM videos . Lots of love and Success Ahead . Can we make any video where input is image and do the RAG. waiting for it.
When we are fetching info from a website, do we need to clean the data as it has many irrelevant words also, or will the model automatically take care of it? Suppose, I am feeding all the pages of a website that is more than 100.....then will it work or need some data cleaning/ pre processing?
Hi Ma’am, your videos are really helpful. Thank you so much for sharing these contents. Through your videos I quickly get to know the new technology coming in the market. But if possible can you hide the api key and in general any keys you use in your videos?
Glad my videos are helpful and I really appreciate your feedback about the API keys. Rest assured, the API keys shown in those videos are deleted after making the videos. But Thanks for pointing that out! 🙂
i had an interview, using openai, i have created a chat bot similar like chatgpt, im able to exactly answer the input question from the document(data) during streamlit, im able to create preview the chatbot, during the each input question the output is not generated fastly, in top right corner their is a option "Running" is previewed then only after 15sec its able to give the answer. because of this, im not able to explain the answer and i lost the job
I'm sorry to hear that your interview didn't go well. Responses delay due to various reasons like - check if you have sufficient resources to run this app. If you're running the demo on a local machine, consider using a machine with a faster CPU or more RAM.
RAG works by first finding relevant information from a large database and then using that information to help generate a better and more accurate response to a question or query asked by user. We can say that RAG is like combining a smart search engine with a powerful text generator.
Hello Arohi-mam , I loved your teaching, and I am trying it out myself. I have one question Which library /steps should i modify if i need to load the contents from a pdf file , instead of URL? your response will be highly appreciated !
It depends - If you want to create detailed text or code responses that pull in information in real time, RAG might be the better choice. But if you need a tool that lets you run code and get explanations interactively, NotebookLM could work better for you.
Very Informative. Thank you.
Glad it was helpful!
I'm searching a good content about RAG for a long time, its very useful to understand about RAG process.
Glad it was helpful!
Clear and to the point . Really like your style of teaching. Learnt quite a bit here. Thank you!!
Glad it was helpful!
Super and easy videos to follow. Keep the good work going.
Thank you so much 🙂
Very easy and perfect way to explain, thanks mam.
Please come up with fine tuning, deployment on cloud, how to test llm model's performance kind of videos.
Your way of explanation is very simple and effective.
@@sagarbhamburkar9395 Sure 👍
Arohi mam, your tutorials are really helpful for me as you explain each and every function, line of code and concept, you are doing great great job
You are most welcome
Really good explanation. Thanks Aarohi!
Welcome :)
Really its good explanation so thanks mam and it is very helpful to students
Glad it is helpful!
Thank you for a detailed and easy-to-understand tutorial.
I have a request, please create a tutorial to implement RAG on any LLM and on any document, such as text files and databases. I'll surely do research from my side, but a help from mentor speeds up the learning process.
I have realized it by now.
Still, thank you for providing the basics, which helped me understand that part
Sure!
have you made the requested video@@CodeWithAarohi ??
Nice explanation as all time
Glad you liked it
Keep the good work.
Thanks, will do!
Wow Nice training you made our life easy. Please do post tutorials regularly
I will try my best
I followed your steps in installation of pytorch and torch vision and open cv yolo v5 etc..success fully I installed opencv2 with cuda support and installed successfully pytorch and torch vision etc but in installation of yolo v5 I got struck at ultralytics then I got know python 3.6 is not compatible with installation of ultralytics. Now I again flashed os into board and I created new virtual environment with python 3.8 now pytorch is not getting installed as my jetpack SDK is 4.5 and I cannot upgrade to 5 as hardware is not supported, but in many articles it is said that ultralytics is important how do I handle this situation 😅
You can run yolov8 with deepstream on Jetson nano. Try that now :) I have a video on that also
Amazing videos mam
Thanks a lot
very good
Thank you!
Hello Aarohi, how did you fetch the url..you didn't explain that part..
Nicely explained
Thank you so much 🙂
Amazing! We need a series playlist from you please.
Working on it!
Thanks for your excellent tutorial, Mam. If possible , can you share some insights to handle hallucination in genAi mam. Thank u.
thank you for the video . I like your consistency in moving with GenAI and LLM videos . Lots of love and Success Ahead . Can we make any video where input is image and do the RAG. waiting for it.
Sure, Will cover the requested topic soon.
When we are fetching info from a website, do we need to clean the data as it has many irrelevant words also, or will the model automatically take care of it? Suppose, I am feeding all the pages of a website that is more than 100.....then will it work or need some data cleaning/ pre processing?
Clean dataset yourself.
Hi , Can you check your Git repo we are unable to see Basics Rag code file
github.com/AarohiSingla/Generative_AI/blob/main/L-7/RAG_demo/basics_RAG.ipynb
Do I need to pay openai to use their api?.
Yes
Can you please explain about the LLM hyper parameters and how it is helpful to get good results
Will cover in my upcoming videos.
Hi Aarohi,
Can you create a video on RAG Implementation on ODA Chatbot where the bot need to interact with the Oracle ADW
I will try.
Hi Ma’am, your videos are really helpful. Thank you so much for sharing these contents. Through your videos I quickly get to know the new technology coming in the market. But if possible can you hide the api key and in general any keys you use in your videos?
Glad my videos are helpful and I really appreciate your feedback about the API keys. Rest assured, the API keys shown in those videos are deleted after making the videos. But Thanks for pointing that out! 🙂
i had an interview,
using openai, i have created a chat bot similar like chatgpt,
im able to exactly answer the input question from the document(data)
during streamlit, im able to create preview the chatbot, during the each input question the output is not generated fastly, in top right corner their is a option "Running" is previewed then only after 15sec its able to give the answer.
because of this, im not able to explain the answer and i lost the job
I'm sorry to hear that your interview didn't go well. Responses delay due to various reasons like - check if you have sufficient resources to run this app. If you're running the demo on a local machine, consider using a machine with a faster CPU or more RAM.
is is RAG just like search key word and provide output ?
RAG works by first finding relevant information from a large database and then using that information to help generate a better and more accurate response to a question or query asked by user. We can say that RAG is like combining a smart search engine with a powerful text generator.
@CodeWithAarohi thanks. Your video showed from website. Can we provide word excel and ppt as input, still can RAG fetch info from those ?
Yes, you can load data from various sources using different data loaders in langchain.
Hello Arohi-mam , I loved your teaching, and I am trying it out myself. I have one question
Which library /steps should i modify if i need to load the contents from a pdf file , instead of URL?
your response will be highly appreciated !
I got the solution
from langchain_community.document_loaders import PyPDFLoader
loader = PyPDFLoader("file.pdf")
data = loader.load()
Check this: github.com/AarohiSingla/Generative_AI/blob/main/L-8/gemini_rag_demo/basics_RAG_pdf.ipynb
Can you please make a vide on how to add chat history for this RAG
Okay
mam please upload few real case oriented LLM projects.
Do you think we can replace RAG by NotebookLM?
It depends - If you want to create detailed text or code responses that pull in information in real time, RAG might be the better choice. But if you need a tool that lets you run code and get explanations interactively, NotebookLM could work better for you.
Thanks ma'am
Most welcome 😊
ma'am plz make the conversational chatbot having previous context too
Sure