OpenAI ChatGPT Embeddings API 🧠 Beginners' EASY Step-by-Step Tutorial 🤯

Key Takeaways

this is amazing is there any way to create embeddings using chat GPT API what is embeddings how it's related to AI That's exactly what we're going to see today let's get [Music] started hi everyone I'm really excited to show you about embeddings and how it's linked to AI not only that I'm going to take you through step by step on how to use chat GPT API in regards to embeding but before that I regularly create videos in regards to artificial intelligence in my YouTube channel so do subscribe and stay tuned so what is embedding it is a process by which you convert text to vectors vectors are nothing but numbers to represent those words it is used for dimensionality reduction that means it simplifies the complex text Data into numerical numbers as you can see man is represented by numbers women is represented by numbers King is represented by numbers and also it creates relationship between words as you can see in the image man woman king queen related together so for a beginner embeddings are nothing but a language for a machine learning model now how it's linked to the overall Ai and also large language models let's take an example as chat GPT large language model if you don't have embeddings chat GPT has its own brain which is limited so when a user asks question without an embedding it can answer only limited number of questions but when you add embeddings as a second brain you can upload unlimited data so when a user ask a question the first step the program checks if there is any information related to those questions in the embeddings or the second brain if there are similar words that will be paed to chat GPT so chat GPT can use information from the second brain and also its own brain to answer the question to make it much more accurate now we're going to see how to create create embeddings and save those embeddings in this tutorial we're going to use chat GPT to create embeddings so first step Conta create hyphen chat GPT python equal 3.11 and then cond activate chat GPT to activate the virtual envirment next export your open AI AP key like this and click enter now we're going to install open Ai and pandas using pip install and click enter now we're going to create a file embedding dopy inside the file import pandas as PD then from open a import open AI then we are initializing open AI client client equals open AI then we are providing example text that is hello world and then we are going to create embeddings client. embeddings doc create function will create the embedding so as a input we are passing the text that is hello world next we are providing the model name that is text embedding add 002 and then we are extracting the embedding finally we are printing the embeddings now I'm going to run this code python embedding dopy and click enter so you can see the the embedding the hello world converted into Vector numbers as you can see in the screen this is called an embedding now we're going to convert this into a function I'm going to Define embeddings so it's the same function we are using client. embeddings doc create so when the text is passed it will automatically return the embedding now we going to print that get embedding hello world it's going to print the same results I'm going to go to the terminal again python embedding dopy and click enter it'll print you the same result we just moved this line inside a function that's it to make our work easy now the next step is that we have a CSV file with product ID user ID score summary and text these are Amazon reviews from thousand products now we are going to feed this data into the second brain that means we going to convert this data into embeddings and store it in a CSV file as you can see I got the CSV file in the same folder as my embeding dop file now we're going to use pandas to read the file pandas is nothing but a CSV file or Excel sheet for developers or data scientists so DF equals pd. read CSV and the CSV file name then we are combining the summary and the text together and saving it in a column called combin this will automatically create a column called combin and merge the summary column and the text column together and put it in the combined column we are going to use only this combined column to convert to embeddings because those are the important data which we want to use and next we are creating embedding column in that embedding column we are going to save all the embeddings so DF do comined is the data which we are going to convert to embedding so we are passing that to the get embedding function this will automatically convert every single Row in the CSV file to a embedding and finally we are saving the embedding in a CSV file now I'm going to run this code in your terminal python embedding dop and click enter now the embeddings are getting generated in the background and it's getting stored in the CSV file now the embeddings got created in the Fine Food reviews with embeddings file that's it as simple as that as a quick summary we collected all the data which we need to convert to embeddings in a column named combined then we pass that data inside the get embedding function and that function will convert those text into embeddings I'm going to create more videos regarding this and also the whole embedding workflow or rag which I will link that in the description below so stay tuned I hope you like this video do like share and subscribe and thanks for watching