@@AlessioDiasparra-lx4ns not sure if you are running the virtual environment. Can you join the Discord server: discord.com/invite/hEMqtDXCHA We can take the discussion further over there
@salesforcecafe1013 күн бұрын
Great work !! Keep helping with your rocking videos !
@ramareddymathsacademy3 күн бұрын
Great video, Tarun! It’s really helpful. Can we have a mobile version where we can scan product details and get reviews/ratings to decide whether to accept or reject the product when I’m at Dmart? The details should be available with just a button click, so I can read them later. This should follow the quality control guidelines set by the Government of India. Just sharing some ideas.👋👍
@AIwithTarun3 күн бұрын
Thank you. Yes, the app is deployed, it can be tested directly on mobile as well. (ingredients-analyzer.streamlit.app/). You need to upload the image and get the results accordingly. Regarding the reviews and ratings, as of now its not implemented, but yes its easy to achieve that along with implementation of quality control guidelines
@subhashchandra33184 күн бұрын
Please make more rag langchain end to end projects with grooq and Gemini api and ollama
@AIwithTarun4 күн бұрын
@@subhashchandra3318 we already have 8-9 videos on that in my channel. But yes project based videos are pending. Maybe January 2nd week
@tora34925 күн бұрын
which IDE are you using & how to get this kind of terminal 17:06?
@AIwithTarun5 күн бұрын
I am using VS Code. For terminal it’s ZSH. When you run print_response you get that kind of results
@digiIntuitions7 күн бұрын
Great video Tarun. Just a fix... the upload image and camera image options in the code will always return wrong output as the streamlit uploader which you are passing to analyze_image() is not the file path rather it is Byte10 class. As a result even you provide the legit image it will always be unable to analyse. We need to temp save the image and then provide the path. This can be done by adding with NamedTemporaryFile(dir='.', suffix='.jpg') as f: f.write(uploaded_file.getbuffer()) analyze_image(f.name) Don't forget --> from tempfile import NamedTemporaryFile A great work BTW, keep rocking.
@AIwithTarun7 күн бұрын
Good catch. Let me fix this asap 😅. Thank you for pointing this out, I will make the changes and try to pin this comment or add description on the changes done in the code. Thank you.
@digiIntuitions7 күн бұрын
@@AIwithTarun Also add "Please do not analyse any other type of images." in the system prompt else it will analyze any type of images. 😊 nothing wrong though, it can be generalised.
@AIwithTarun7 күн бұрын
The app is updated: github.com/lucifertrj/Product-Ingredient-Agent/ Thank you :)
@digiIntuitions7 күн бұрын
@@AIwithTarun Great! Off topic, what software you use to create videos? I want to create few videos as well, if you don't mind telling me.
@AIwithTarun7 күн бұрын
@@digiIntuitions sure. I use iPhone to capture the face, QuickTime Player to screen record and finally Final Cut Pro to edit or merge the video. Initially I was using Zoom.
@nikith157 күн бұрын
Do u provide ai agent building services?
@AIwithTarun7 күн бұрын
@@nikith15 as of now no. Maybe from next month or February
@stalinmurugesan97447 күн бұрын
Thank you very muchTarun. when I try first approach I get only few line message like "The image shows a product package for Bournvita, a nutritional supplement. The package is primarily orange and white,etc. Not full details like your output. Can you let me know wny ? I used same System_prompt, Instructions,etc.
@AIwithTarun7 күн бұрын
Interesting. Can you provide temperature = 0 and may I know what LLM are you using?
@stalinmurugesan97447 күн бұрын
@@AIwithTarun Thank you for your prompt response. I used Gemini Flash 2.0..
@AIwithTarun7 күн бұрын
@ can you add temperature = 0 and rerun the code.
@SuhasB-ke6mu7 күн бұрын
I want to create a LocalRAG system (chat with PDF) using Llama 3.2 and text embeddings. However, the results often include hallucinated information. Do you have any suggestions on how to train and test the model to ensure the system provides accurate answers?
@AIwithTarun7 күн бұрын
There are various factor to check with you are working on RAG using Open Source LLMs: - Have you used the Prompt template as used in Llama3.2? If your context is getting extracted you need to augment your prompt to reduce hallucinations [However this is not 100% accurate but it reduces the risk] - On the retriever part, you need to check if the relevant documents is retrieved or not for the user query. This is where you need to try CRAG or Re-reranking to improve the performance. You can join our Discord server, we can take this discussion further to see where things are going wrong.
@AIwithTarun7 күн бұрын
Here is my repo: github.com/lucifertrj/Awesome-RAG/ I have most of the colab notebook that uses Open Source LLM itself.
@ArpitSingh-wp6yx7 күн бұрын
Bro any vdo for building agents for CRM works ?
@AIwithTarun7 күн бұрын
I haven’t planned for it. Need to think about it. Meanwhile if you have any questions on building it, join our Discord channel, we can have discussion over there
@תמרכהן-כ4ק9 күн бұрын
Thank you for this series! You are a great teacher 🫶
@AIwithTarun9 күн бұрын
Thank you🚀 We are just getting started. It’s only been 3 videos yet. More videos on the way. Keep supporting
@ghulamjunudchishti34409 күн бұрын
Thank you brother, I always watch your videos.
@AIwithTarun9 күн бұрын
Thank you brother. Keep supporting and watch the videos. I hope you build some cool projects with this 🚀
those 3 organization priority is something I will definitely try next year. Thank you
@AIwithTarun11 күн бұрын
@@RashomonAI I am glad you found it useful. Good luck 🚀
@piyushsahu59112 күн бұрын
Please come to mumbai for an event 🙏
@AIwithTarun12 күн бұрын
@@piyushsahu591 Mumbai is already in timeline. Dates needs to be confirmed
@piyushsahu59110 күн бұрын
@AIwithTarun okay brother
@AlphaGodz13 күн бұрын
Hi Amazing explanation about agentic tool calls from different approaches! 👍 I Have 2 questions 1. May I know what dashboard are you using to create the flow diagrams ? 2. Can you make a video on Event driven llama-index Agentic workflows ?
@AIwithTarun13 күн бұрын
@@AlphaGodz Thank you for the kind words. - I either use Excalidraw or Draw.io - Sure. Do join our Discord for further discussions and reminder for the video. Link to join is in description
@lohithArcot14 күн бұрын
How does it compare with Claude?
@AIwithTarun14 күн бұрын
Claude is a LLM. AIDE is an IDE. Moreover internally AIDE is using Claude for the assitant chat feature. (this is configurable: you can pick Claude, Deepseek and other models it support)
@reserseAI11 күн бұрын
🤦
@tym_pass21 күн бұрын
I tried myself just now.. this AI is not that vast and clever. Specifically not creating good images for most Indian well known person.. I also tried imaginary prompt like cat riding a bicycle.. it just pasted a cat face on a guy on bicycle.. my prompt was professional headshot portrait of Cat riding a bicycle, extremely detailed, ultra realistic, 8k, shoot with canon EOS r5
@AIwithTarun21 күн бұрын
- As mentioned in the video, keyword is very important. Your prompt is logically wrong, you have mentioned the keyword Professional Headshot for Cat, which is logically incorrect for Animals. Try this instead: ``` Photograph a {{cat}} riding a {{bicycle}} in a sunny park, the cat is positioned upright with its front paws on the handlebars and appears to be in motion, natural light, vibrant colors, ultra detailed, {{{realistic}}}, photographed on a Canon EOS R5, 50mm lens ``` - The image output depends on keywords, use it wisely. - If you watched the video, I tried with Virat Kohli, it didn't give the exact photo as needed, that means model have no data to make it realistic. In this case you need to use InstantID model. Furthermore, in Recraft, when you click on the generated image there is also Fine Tune option, currently I am experimenting with it, will make the video on that soon.
@playloop883222 күн бұрын
Very informative. Thank you!
@changtimwu23 күн бұрын
came from qdrant's X repost. The terminal output is so fancy. Maybe make it as a RAG CLI. Just search among personal frequently used knowledge bases.
@AIwithTarun19 күн бұрын
You can use this: agent.cli_app(stream=True) instead of running print_response.
@phidata29 күн бұрын
Thank you for featuring Phidata, Tarun !!
@revanthbhuvanagiri9177Ай бұрын
Hey , really excellent work, I have issue while openAI API keys , could you please make video on how to create openAI api keys , it would be really helpful.😁😄
@AIwithTarunАй бұрын
- Go to platform.openai.com and login - Click on dashboard and navigate to API keys - Create a new key and save it somewhere. Also to remind you OpenAI API keys are not free, if you have the access then you can use it. If not use Gemini or XAI (grok)
@anshumanngupta5030Ай бұрын
you could hv explained better with n8n
@AIwithTarunАй бұрын
@@anshumanngupta5030 I have not used n8n before. Will check it out once
@manojkurdekar2400Ай бұрын
Your tutorial are definitely better compared to others I've seen especially with RAG maintenance in general. Thanks
@mihir5846Ай бұрын
Who exactly is making a decision whether to use the knowledge base or tool, openai model? Can a similar thing work with any other model? And how do you make a custom tool and decision making?
@AIwithTarunАй бұрын
LLM is doing the decision making part with heavy prompting. It depends on model reasoning capabilities. Gpt4-o, Claude, Grok XAI have good reasoning capabilities. Custom tool logic is possible, it just a function that returns context. Maybe I will make one video on that.
@mihir5846Ай бұрын
@@AIwithTarun Thanks for the reply. Looking forward to your future videos.
@harshasr3705Ай бұрын
My fav sensei back at it making things easy for me!!! Love your content
@RashomonAIАй бұрын
Damn, this is soo cool. The inference and quality looks promising. As always loved the flow of video. Thanks for sharing
@juangonzalezcabello1737Ай бұрын
Good job, subscribed and shared! I have some thougths about this, reached you via linkdin to have a chat!
@aiplanet2 ай бұрын
Cool
@zavtherave2 ай бұрын
how to get 9.5 gpa like u great master
@AIwithTarun2 ай бұрын
GPA doesn't matter. build projects and contribute to Open Source
@Koala-muncher2 ай бұрын
Amazing video, keep up the good work
@digitaldepot232 ай бұрын
I need an AI artist who knows Lora's training. I will pay. Is anyone here to help me?
@Alex-dr6or2 ай бұрын
I can help
@digitaldepot232 ай бұрын
@@Alex-dr6or Contact details
@Premanandbaba-h7b26 күн бұрын
How much u can pay
@huythai8552 ай бұрын
Hi! Thank you for your very easy-to-understand video. May I ask which program or website you used to draw the execution flow in the video?
@AIwithTarun2 ай бұрын
Thank you. I use: excalidraw.com/
@Mrajayreddy2 ай бұрын
Big fan 🔥
@manojkurdekar24002 ай бұрын
Good to know the insight for the GSoC
@arturgoraus79472 ай бұрын
Nice approach. I see most of the people are using LlamaIndex Workflows for Agentic AI/RAG. What is the difference for using ReActAgent and AgentRunner by you? Is Workflows not better supported? I'm asking as I'm planning to develop custom based Chatbot using internal data.
@AIwithTarun2 ай бұрын
@@arturgoraus7947 Agent Runner is mainly dependent on Function calling and LLM, there is no feedback mechanism. Whereas React based Agents are pretty good in error handing and feedback. You can pick React Agent over Agent Runner
@arturgoraus79472 ай бұрын
@@AIwithTarun Thank you. Could you also share your opinion about LlamaIndex Workflows, please? Are you maybe planning to do a video about it? Am I correct by saying that LlamaIndex Workflows is the most supported among LlamaIndex tools? It's a bit strange to me that LlamaIndex has so many different tools to deploy Agentic RAG, making it difficult to pick one.
@AIwithTarun2 ай бұрын
@@arturgoraus7947 Yes, I went through the LlamaIndex Workflows, I will most probably include that in Awesome RAG GitHub repo instead of video. I am planning for project based videos currently.
@arturgoraus79472 ай бұрын
@@AIwithTarun Thank you. That would be useful :)
@dibyajyotiacharya89162 ай бұрын
Great work ❤
@kurtcobain6412 ай бұрын
I heard a lot good things about SearXNG as a web search tool. Hope you make an video following that.
@agirlnamedsrishti3 ай бұрын
After successfully installing lama index , " ModuleNotFoundError: No module named 'llama_index.multi_modal_llms.generic_utils'" error is showing . Please help me resolve it
@AIwithTarun3 ай бұрын
this is mainly because of dependency issues. Can you join the Discord server and address this issue. You can also share the colab notebook there.
@jasonsheinkopf3 ай бұрын
This video is very cool. I'd like to deploy an app for my portfolio. Is there a good method to limit the usage to manage your API costs? Can you set a monthly/daily limit for API? Also, is it possible to put a password on a Streamlit app that you can include with your resume?
@AIwithTarun3 ай бұрын
Regrading the limits, I believe you need to set this at the dashboard end when you are creating the API key. As per the password, yes in Streamlit you can build the Authenticator to add the password, if that is what you are looking for.
@sudiptadas37463 ай бұрын
Amazing, thank you.
@mohammad-xy9ow3 ай бұрын
plzz make a llamaindex and qdrant hybrid search video
@AIwithTarun3 ай бұрын
Sure. Based on the comments: I’ve added a new notebook that covers two implementation: - Qdrant Hybrid Search - Memory integration for the Index engine Check `Qdrant Hybrid Search + Memory` colab notebook in Awesome RAG GitHub repository: github.com/lucifertrj/Awesome-RAG
@limjuroy70783 ай бұрын
How can I make it conversational where the chatbot carries the chat history context? For example: """ 1st ques: Who won the basketball gold medal for men team in the Tokyo Olympics 2020? 2nd ques: How many times that the team have won the gold medal so far in the Olympics? """ According to the example questions above, the Chatbot should be able to know "the team" in the second question is referring to the name of the team that won the gold medal in the Tokyo Olympics 2020.
@AIwithTarun3 ай бұрын
in order to achieve this, you need to integrate chat buffer memory component within your pipeline.
@limjuroy70783 ай бұрын
@@AIwithTarun I see. Maybe you can include this in your coming tutorials.
@AIwithTarun3 ай бұрын
@@limjuroy7078 Sure noted. Before that I will try to push the code here: github.com/lucifertrj/Awesome-RAG to save time
@limjuroy70783 ай бұрын
@@AIwithTarun Thanks!
@AIwithTarun3 ай бұрын
@@limjuroy7078 Based on the comments: I’ve added a new notebook that covers two implementation: - Qdrant Hybrid Search - Memory integration for the Index engine Check `Qdrant Hybrid Search + Memory` colab notebook in Awesome RAG GitHub repository: github.com/lucifertrj/Awesome-RAG