Literaly nobody is actualy showing how to setup the environment and correct order of installing the packages. Its a fact that you wil always get a llama_cpp and chromda db dependencies error with langchain. Nobody on internet is correctly addressing this. If it was just about executing the code it would have been a dreamland, but it isnt. Request you to actualy show how to setup from scarch.
ok so everything is running achese but apparently my pdf is not being useed like the bot isnt answering frm the pdf / when i ask tell me bout the pdf it says provide details ya i cant use the pdf
I am getting this error while giving any query , i have installed the ollama and also downloaded llama3 Error raised by inference API HTTP code: 404, {"error":"model \"llama3\" not found, try pulling it first"}
I fixed an issue with this code. The code works better if you don't have that if condition for uploading pdf files. Your code sometimes doesn't read the pdf file and the model hallucinates. I used Metadata comparison to handle the files better.
Hi, the pdf file is being uploaded and vectordb is being created yet when i ask the model about my document it cant know which document im referring to.
This is why i hate UA-cam, your video should really have gained huge views but no.. shitty videos will gain and push your down.. hate to see this keep up the good work and don't give up!
Hi Thanks for the comment! You'll have to use embeddings or llms to fine tune the model with your data, rag is a retreival framework which gets data and augements it, fine-tuning is training a model on your data, in which case you would not need RAG.
Literaly nobody is actualy showing how to setup the environment and correct order of installing the packages. Its a fact that you wil always get a llama_cpp and chromda db dependencies error with langchain. Nobody on internet is correctly addressing this. If it was just about executing the code it would have been a dreamland, but it isnt. Request you to actualy show how to setup from scarch.
i'll take a look
ok so everything is running achese but apparently my pdf is not being useed like the bot isnt answering frm the pdf / when i ask tell me bout the pdf it says provide details ya i cant use the pdf
I am getting this error while giving any query , i have installed the ollama and also downloaded llama3 Error raised by inference API HTTP code: 404, {"error":"model \"llama3\" not found, try pulling it first"}
I fixed an issue with this code. The code works better if you don't have that if condition for uploading pdf files. Your code sometimes doesn't read the pdf file and the model hallucinates. I used Metadata comparison to handle the files better.
can you share the file or github repo please ?
Yes I'll send it to your email
Hey I have tried your code. The pdf I uploaded is not being saved and the chat is unable to read the uploaded file. How do I fix this error ?
Did you figure this out yet?
Hi, the pdf file is being uploaded and vectordb is being created yet when i ask the model about my document it cant know which document im referring to.
This is strange, please remove both folders and retry
its great job. thanks
This is why i hate UA-cam, your video should really have gained huge views but no.. shitty videos will gain and push your down.. hate to see this keep up the good work and don't give up!
Thanks man! We'll get there
Do you have idea how to fine tune my pdf data with rag ?
Hi Thanks for the comment! You'll have to use embeddings or llms to fine tune the model with your data, rag is a retreival framework which gets data and augements it, fine-tuning is training a model on your data, in which case you would not need RAG.
Can it do accessibilty Check as pdf pro dc does for pdf documents?
you can write custom code for it, yes
@@goyashy can u provide steps n explain in details
So I can share with my friends and ubwill get subs n like n shares
Please explain how