How To Get Answers From Your PDF Using Azure OpenAI And Langchain

Check out this video to learn about querying your PDF file using Azure OpenAI and Langchain.

** REFERRAL LINK *************
** REFERRAL LINK *************

###### MORE PLAYLISTS ######

#openai #chatbot #chatgpt

[ebjwmzf]

Excellent!! video but can you / someone please guide us on how follow up on the next steps to actually host this (on Azure as a web app), so this can be showcased as fully working prototype, as most of the tutorial videos on YT just end up on how to use just within the IDE. Thanks

[trwbdhs]

I am getting an error on Chroma.from_documents
InvalidRequestError: Must provide an 'engine' or 'deployment_id' parameter to create a
Can you please help?

[benjoaoc.]

Amazing! Thank you for the video. I'm trying to split a book into chapters for a small project using OpenAI. Does anyone in the community have a solution?

[ngates]

Thank you so much !!! Have been searching it for long

[nepalsudip]

Also, I think you are using VSC on local environment. How do we use the code inside Azure environment. Is there anything like sagemaker, where I go to Azure and type in code there instead of local python file? Sorry, I am new to Azure. And your videos are great!

[ajitha]

@shweta I had a azure qna maker which I integrated with azure open AI, can I integrated this with teams app ?

[AIenthu]

Thank you so much. Please also provide link to the code, it would be really very helpful

[sanishthomas]

nice. quick question, if I am missing Azure Subscript then where I can use this coding and how Can integrate this with Azure Applicarion

[iranmike]

Thank you. I got this error: The API deployment for this resource does not exist. I wait for days but still there is this error

[prateekagnihotri]

Hi Shweta, Wonderful explanation, I implemented same code, but getting following error in the last line "chain.run(query)"- TypeError: missing 1 required positional argument: 'param', can you please guide

[nepalsudip]

Do you have notebook and input file? Seems like I can't replicate it. Also, it would be helpful how do you set up your visual studio code project.

[malexandersalazar]

I didn't know about that Chroma class. I usually used instead of Chroma.from_documents. It seems that Chroma works from memory, while GPTVectorStoreIndex can work from physical files on disk.

[MotivateMania]

@shweta can you cover different version using gpt and embedding gateways instead of api key

[mayureshbhola]

I am getting an error on Chroma.from_documents
InvalidRequestError: Must provide an 'engine' or 'deployment_id' parameter to create a....

Are the models used for langchain are different what we have for open ai, I have tried multiple models : davinci, 35 turbo etc, but nothing worked out for me. Can you please elaborate on that part ?

[sanishthomas]

what is the console where you are writing lanchain code? Also how do you import the libray to use this langchain

[shripadbhargave]

What plugin of VS code you are using in this video? Looks similar to Jupyter notebooks

[jgnxxmh]

great. what if we have data in multiple PDF files, excel sheet and in csv file. how to get answer from all of this format files data.
is there any way to build this?

[sumangautam]

Can you explain, if I have a pdf file stored in blob storage or datastore, how this works?

[harshsinha]

The embeddings are not
it shows incorrect api keys for Openai which is to be set from platform.openai

[pragmaticprogrammer]

@shweta-lodha Thanks for quick and intresting one. Im struck while calling chroma.from_documents, getting an error "The embeddings operation does not work with the specified model, text-davinci-002." .If possible can you help me with this issue. Thanks ina dvance