🙏❣join to my Patreon: www.patreon.com/GaoDalie_AI Thank you so much for watching guys! I would highly appreciate it if you Book an Appointment with me: topmate.io/gaodalie_ai Support the Content (every Dollar goes back into the video): buymeacoffee.com/gaodalie98d Subscribe Newsletter for free: substack.com/@gaodalie FOLLOW ME : join my discord if you have any questions: discord.gg/GENrSVJN Follow me on Twitter: twitter.com/mr_tarik098 Follow me on Linkedin: shorturl.at/dnvEX Follow me on Medium: medium.com/@GaoDalie_AI More Ideas On My Page: quickaitutorial.com/th-cam.com/users/sgaming/emoji/7ff574f2/emoji_u2763.png
Dspy, ARAG & latest fusion was ultimate. I built it together based on your guidance & it works amazing. I have a small doubt. Isn't graph rag to be done specifically with Neo4j types Graph DBs or how is it, Can u make a video with Neo4j for better understanding
@@GaoDalie_AISaw Graph RAG too again. I thought Graph can only be stored in Graph DBs like Neo4j. But Rag FUSION+CRAG data is stored in chromadb which is not a Graph DB. This confused me now. I loved the implementation & ur efforts in both the methods. But confusion strikes. How chromadb handles Graph
@@StnImg Yes, that is true. In Fusion + CRAG, I use ChromaDB; I didn't use Graph DB. I just understood your question, and I apologize for the misunderstanding. Yes, you can only use Neo4j because their data is stored on their cloud. We just hit the API to extract data. I hope I could answer your question
Boss, i finally understood your implementation after understanding indepth on GraphDB. Your implementation is more of a procedure oriented & if I have to persist the nodes, relationship & proprties, I need to employ Neo4j DB. I added Neo4j, modified chroma to Qdrant to see my embedding & employed crew ai agents. Also added Redis caching & celery Message broker & added enhancements to make it an adaptive RAG with my trained NLP model as one more layer to refine. I'm getting the best of the best results.🎉🎉🎉
I'm trying to understand the use case for this - in most situations where you are orchestrating across LLMs, you will have enough consistency in query structure that you probably would have more success leveraging traditional NLP and business rules for routing. While the benefits of the LLM learning to improve its routing from user patterns seems useful at first glance, that's still a lot of inconsistency we desperately want to avoid when making these decisions, lest we send "code me a website showing our menu" to the 4B "food/menu expert" instead of the "web dev" expert even 1% of the time. Maybe I lack the imagination for where an LLM is better here? Because the only situation I can imagine is for facing a generalized userbase with generalized queries, which none of us are likely ever going to compete with :D
Hey, thank you for reaching out. Yes, I can assist you if you subscribe to Rainbow Superstar or Diamond Superstar. I also have a Discord for free users where I share information, but I do not provide assistance or help there.
Thank you for watching. I don't know where I created misinformation, but if you look carefully, you will see I highlighted two models: a weak model and a strong model, which are two LLMs you can choose from. Thanks again, and welcome to join us
Great Video Nice
Thank you for watching
🙏❣join to my Patreon: www.patreon.com/GaoDalie_AI
Thank you so much for watching guys! I would highly appreciate it if you
Book an Appointment with me: topmate.io/gaodalie_ai
Support the Content (every Dollar goes back into the video): buymeacoffee.com/gaodalie98d
Subscribe Newsletter for free: substack.com/@gaodalie
FOLLOW ME :
join my discord if you have any questions: discord.gg/GENrSVJN
Follow me on Twitter: twitter.com/mr_tarik098
Follow me on Linkedin: shorturl.at/dnvEX
Follow me on Medium: medium.com/@GaoDalie_AI
More Ideas On My Page: quickaitutorial.com/th-cam.com/users/sgaming/emoji/7ff574f2/emoji_u2763.png
Dspy, ARAG & latest fusion was ultimate. I built it together based on your guidance & it works amazing. I have a small doubt. Isn't graph rag to be done specifically with Neo4j types Graph DBs or how is it, Can u make a video with Neo4j for better understanding
hey, thank you for reaching out, please I have made a video about Graph Rag you can check my videos in my profile.
@@GaoDalie_AISaw Graph RAG too again. I thought Graph can only be stored in Graph DBs like Neo4j. But Rag FUSION+CRAG data is stored in chromadb which is not a Graph DB. This confused me now. I loved the implementation & ur efforts in both the methods. But confusion strikes. How chromadb handles Graph
@@StnImg Yes, that is true. In Fusion + CRAG, I use ChromaDB; I didn't use Graph DB. I just understood your question, and I apologize for the misunderstanding. Yes, you can only use Neo4j because their data is stored on their cloud. We just hit the API to extract data. I hope I could answer your question
Boss, i finally understood your implementation after understanding indepth on GraphDB. Your implementation is more of a procedure oriented & if I have to persist the nodes, relationship & proprties, I need to employ Neo4j DB. I added Neo4j, modified chroma to Qdrant to see my embedding & employed crew ai agents. Also added Redis caching & celery Message broker & added enhancements to make it an adaptive RAG with my trained NLP model as one more layer to refine. I'm getting the best of the best results.🎉🎉🎉
@@StnImg bro you are a rock star , i appreciate your hardworking , would you like to share the code with me to take a look how you did it thanks
I'm trying to understand the use case for this - in most situations where you are orchestrating across LLMs, you will have enough consistency in query structure that you probably would have more success leveraging traditional NLP and business rules for routing. While the benefits of the LLM learning to improve its routing from user patterns seems useful at first glance, that's still a lot of inconsistency we desperately want to avoid when making these decisions, lest we send "code me a website showing our menu" to the 4B "food/menu expert" instead of the "web dev" expert even 1% of the time.
Maybe I lack the imagination for where an LLM is better here? Because the only situation I can imagine is for facing a generalized userbase with generalized queries, which none of us are likely ever going to compete with :D
hello quick question will you guide me with resources if i patreon for your discord? i couldnt run the model i bought(my fault)
Hey, thank you for reaching out. Yes, I can assist you if you subscribe to Rainbow Superstar or Diamond Superstar. I also have a Discord for free users where I share information, but I do not provide assistance or help there.
This video basically creates mis-information. Not for practitioners. RouteLLM doesn't really do what they suggest. You can only choose between 2 llms.
Thank you for watching. I don't know where I created misinformation, but if you look carefully, you will see I highlighted two models: a weak model and a strong model, which are two LLMs you can choose from. Thanks again, and welcome to join us