1 00:00:00,110 --> 00:00:06,800 ‫ChatGPT is an advanced natural language processing model that uses deep learning techniques to understand 2 00:00:06,800 --> 00:00:08,870 ‫and generate human like texts. 3 00:00:09,800 --> 00:00:16,910 ‫In simple words, if we ask anything to tell GPT, then it will provide us with an answer of our question, 4 00:00:16,910 --> 00:00:21,710 ‫and that generated text sound like it's coming from a human. 5 00:00:22,160 --> 00:00:26,540 ‫Now let's discuss how we can integrate ChatGPT with Jupyter Notebook. 6 00:00:26,540 --> 00:00:33,830 ‫So by using the open AI Python package, we can connect our Jupyter notebook environment to the ChatGPT 7 00:00:33,830 --> 00:00:34,430 ‫API. 8 00:00:34,940 --> 00:00:37,430 ‫To get started with the ChatGPT API. 9 00:00:37,460 --> 00:00:40,130 ‫First of all, you need an OpenAI account. 10 00:00:40,130 --> 00:00:44,240 ‫If you don't have one, then you have to sign up in your OpenAI account. 11 00:00:44,240 --> 00:00:53,000 ‫For this, you have to visit OpenAI website, which is openai.com, and here you have to log in or sign 12 00:00:53,000 --> 00:00:53,960 ‫up to your account. 13 00:00:54,200 --> 00:00:57,650 ‫Once you have created your account, navigate to the OpenAI dashboard. 14 00:00:57,650 --> 00:01:02,180 ‫You can usually find dashboard or account link in the top right corner of the website. 15 00:01:03,020 --> 00:01:05,060 ‫Click on it to access your dashboard. 16 00:01:05,990 --> 00:01:10,910 ‫In the dashboard, you will find various sections and settings related to your OpenAI account. 17 00:01:11,360 --> 00:01:17,750 ‫Look for the API keys or credentials section as this is where you can generate your API key. 18 00:01:17,930 --> 00:01:22,010 ‫Click on the API keys or credentials section to expand it. 19 00:01:22,010 --> 00:01:25,070 ‫Here you will have the option to create a new API key. 20 00:01:25,730 --> 00:01:30,680 ‫Click on the Create New Key or Generate Key button to generate your API key. 21 00:01:31,460 --> 00:01:34,250 ‫Once you have generated your API key, copy it. 22 00:01:34,250 --> 00:01:39,680 ‫This key is essential for authenticating your request to the ChatGPT API. 23 00:01:40,130 --> 00:01:41,930 ‫That's how we log in to OpenAI. 24 00:01:42,050 --> 00:01:45,500 ‫Create an account and obtain your API key. 25 00:01:46,100 --> 00:01:51,680 ‫Now let's move on to the next step integrating the ChatGPT API with Jupyter Notebook. 26 00:01:53,900 --> 00:02:00,830 ‫With the Openeye Python package and a few lines of code, we can connect our Jupyter Notebook environment 27 00:02:00,830 --> 00:02:06,080 ‫to the ChatGPT API to integrate the ChatGPT API with Jupyter Notebook. 28 00:02:06,110 --> 00:02:09,350 ‫We will be using the OpenAI Python package. 29 00:02:09,380 --> 00:02:12,590 ‫First, make sure that you have the package installed. 30 00:02:12,590 --> 00:02:18,110 ‫You can do this by running pip install OpenAI in your Jupyter notebook environment. 31 00:02:18,380 --> 00:02:22,370 ‫Which code I am using here is stored in this API file. 32 00:02:22,370 --> 00:02:26,450 ‫I have provided a link of this file in the description of this video. 33 00:02:26,630 --> 00:02:30,920 ‫If you want to use this code, then you have to download this code and upload it. 34 00:02:30,920 --> 00:02:35,450 ‫From here we will open this file and now we have our codes ready. 35 00:02:35,480 --> 00:02:38,120 ‫Let me explain you this code for you. 36 00:02:38,120 --> 00:02:42,890 ‫So here you have to replace this API key with your actual API key. 37 00:02:43,100 --> 00:02:49,850 ‫And then we have a function which is get GPT response that takes the prompt and the optional max token 38 00:02:49,850 --> 00:02:50,570 ‫parameter. 39 00:02:51,110 --> 00:02:57,350 ‫This function will communicate with ChatGPT API and return the response based upon the given prompt. 40 00:02:57,380 --> 00:03:01,490 ‫Next we make the API response using completion create. 41 00:03:01,490 --> 00:03:09,500 ‫Here our model is GPT 3.5 turbo and then message max tokens and the temperature. 42 00:03:10,280 --> 00:03:14,300 ‫Experimenting with the temperature can achieve different styles of response. 43 00:03:14,300 --> 00:03:16,910 ‫And at last our response. 44 00:03:17,840 --> 00:03:19,490 ‫Now we will write our prompt. 45 00:03:20,270 --> 00:03:24,680 ‫So first I will write what is ChatGPT? 46 00:03:25,580 --> 00:03:26,720 ‫Let's run this code. 47 00:03:28,470 --> 00:03:34,230 ‫And here you can see it has written that ChatGPT is a language model developed by OpenAI. 48 00:03:34,260 --> 00:03:40,260 ‫It is based on the GPT three architecture, which uses deep learning techniques to generate human like 49 00:03:40,260 --> 00:03:42,750 ‫text responses and so on. 50 00:03:43,020 --> 00:03:50,910 ‫Now I will write my second prompt, which is how to treat outliers in machine learning. 51 00:03:51,540 --> 00:03:59,010 ‫If I run this code here, you can see that it has actually given me four ways to treat outliers, which 52 00:03:59,010 --> 00:04:03,240 ‫is removal, miniaturization, transformation, and binning. 53 00:04:03,750 --> 00:04:07,710 ‫In the next prompt, I will write What is Game of Thrones? 54 00:04:07,920 --> 00:04:09,180 ‫Let's run this code. 55 00:04:10,440 --> 00:04:15,930 ‫And here you can see that the Game of Thrones is a popular fantasy television series based on the book 56 00:04:15,930 --> 00:04:19,410 ‫series A Song of Ice and Fire by George Martin. 57 00:04:19,410 --> 00:04:25,650 ‫So you can see that with the integration, we can use the power of ChatGPT within our Jupyter notebook. 58 00:04:25,650 --> 00:04:27,180 ‫So that's all in this video. 59 00:04:27,210 --> 00:04:28,470 ‫See you in the next one. 60 00:04:28,470 --> 00:04:29,130 ‫Thank you.