QA chat with a website, OpenAI embeddings tutorial: Use GPT 3 API and Openai ADA-2 Embeddings

We will learn about how to convert the text contents of a website in to OpenAI embeddings to be able to do similarity search for context for our question answering GPT 3 API.

timeline:
00:00 demo
00:37 tutorial overview
02:11 Creating a conda environment
04:10 Pip installing requirements txt
05:45 Following the tutorial step by step
10:39 Token amount considerations
11:45 creating the embeddings
13:00 Asking questions
14:25 overall review

#openaiembeddings #openai #openaitutorial #gpt3 #gpt #gpt3tutorial

[KeshavMelnad]

Very useful video. Thank you. I just have a question. When it finds the correct context from embeddings and appends to the question and then try to answer, the total number of tokens in the prompt will the our question + context right? In that case, the cost of the question-answering increases. Am I right? Is there a way to minimise cost?

[jaysonp]

This is great but I don't see anywhere in the code where we're passing our OpenAI key

[shivamkumar-qpjm]

From some days I watch this channel as a Netflix shows but unfortunately Indian card is not working unable to purchase

[echohive]

I hope You will find this video helpful! :)

[kajasheriff]

If I use my website, it will not get the information??

[mvrdara]

this can distrupt algolia totally! which is a multi billlion dollar company