1 00:00:00,000 --> 00:00:01,000 Hello guys! 2 00:00:01,000 --> 00:00:07,000 So finally, I'm excited to implement the first end to end gen AI application that is nothing but a 3 00:00:07,000 --> 00:00:08,000 Q&A chat bot. 4 00:00:09,000 --> 00:00:11,000 Uh, I'll just give you a brief architecture. 5 00:00:11,000 --> 00:00:15,000 What all things we are specifically going to do in this particular project. 6 00:00:15,000 --> 00:00:18,000 And then from the next video we will start implementing it. 7 00:00:18,000 --> 00:00:24,000 So over here, uh, some of the tools or some of the APIs that or some of the LLM models that we are 8 00:00:24,000 --> 00:00:27,000 going to use is OpenAI LLM models. 9 00:00:27,000 --> 00:00:27,000 Okay. 10 00:00:27,000 --> 00:00:36,000 So we are going to use OpenAI LM models, uh, to name some of them are nothing but GPT four. 11 00:00:36,000 --> 00:00:41,000 So this is uh this is a multi model okay. 12 00:00:41,000 --> 00:00:42,000 GPT four. 13 00:00:42,000 --> 00:00:45,000 Then we will also be using GPT four turbo. 14 00:00:45,000 --> 00:00:46,000 Right. 15 00:00:46,000 --> 00:00:51,000 We will create this entirely in a Streamlit web app so that you'll have multiple options to select the 16 00:00:51,000 --> 00:00:53,000 model and probably continue it. 17 00:00:53,000 --> 00:00:54,000 Okay. 18 00:00:54,000 --> 00:00:57,000 So this is one of the LM models. 19 00:00:57,000 --> 00:01:00,000 The second one that we are going to probably use is llama. 20 00:01:00,000 --> 00:01:05,000 Now with the help of llama, what we are going to do is that whatever open source models are available 21 00:01:05,000 --> 00:01:06,000 in llama. 22 00:01:06,000 --> 00:01:06,000 Right. 23 00:01:07,000 --> 00:01:10,000 Like uh, llama two. 24 00:01:10,000 --> 00:01:10,000 Sorry. 25 00:01:10,000 --> 00:01:13,000 Llama three, gamma two. 26 00:01:13,000 --> 00:01:16,000 So here we are just going to write gamma two. 27 00:01:16,000 --> 00:01:19,000 Along with that, if you are also familiar with Mistral. 28 00:01:19,000 --> 00:01:24,000 So all these open source models also will try to use and create this Q&A chatbot okay. 29 00:01:24,000 --> 00:01:29,000 So in our project, first of all, what we are going to do is that we are going to create the Streamlit 30 00:01:29,000 --> 00:01:30,000 web app. 31 00:01:31,000 --> 00:01:37,000 So this will be my Streamlit web app. 32 00:01:38,000 --> 00:01:41,000 This Streamlit web app will be interacting. 33 00:01:41,000 --> 00:01:48,000 I'll give an option wherein it will be interacting with the OpenAI APIs. 34 00:01:49,000 --> 00:02:00,000 Okay, open AI, OpenAI API, and with the help of this OpenAI API, we will try to interact with various 35 00:02:00,000 --> 00:02:09,000 LLM models like GPT four or GPT four turbo, or it can also be GPT. 36 00:02:10,000 --> 00:02:11,000 GPT. 37 00:02:11,000 --> 00:02:13,000 Let's say for over here. 38 00:02:13,000 --> 00:02:14,000 Right. 39 00:02:14,000 --> 00:02:21,000 Once we interact with this, then you'll be able to see that we will finally get our response. 40 00:02:22,000 --> 00:02:25,000 Now we're not going to stop over here right? 41 00:02:25,000 --> 00:02:33,000 Once this is entirely interacting with this LLM models, we'll take one step ahead and we will further 42 00:02:33,000 --> 00:02:39,000 put all these logs in our Lange Smith platform. 43 00:02:39,000 --> 00:02:39,000 Okay. 44 00:02:39,000 --> 00:02:41,000 So Lange Smith will also be there. 45 00:02:42,000 --> 00:02:46,000 So here you will be able to see that we will be having Lange Smith. 46 00:02:46,000 --> 00:02:48,000 Now Lange Smith right is used. 47 00:02:49,000 --> 00:02:52,000 It is specifically used for monitoring. 48 00:02:54,000 --> 00:02:58,000 Then it is used for debugging, right? 49 00:02:58,000 --> 00:03:01,000 To also see how much cost is basically getting involved. 50 00:03:01,000 --> 00:03:06,000 So all this kind of information will be available in the blacksmith once you are able to track it. 51 00:03:06,000 --> 00:03:09,000 So here we are just going to give our user query. 52 00:03:10,000 --> 00:03:12,000 Then all this process is going to happen. 53 00:03:12,000 --> 00:03:18,000 Okay I know this looks like a very simple project, but understand one thing is that as we go ahead, 54 00:03:18,000 --> 00:03:21,000 we'll be building more complex project on top of it, right? 55 00:03:21,000 --> 00:03:22,000 Like Rag applications. 56 00:03:23,000 --> 00:03:26,000 Then we will be seeing Rag with SQL. 57 00:03:26,000 --> 00:03:29,000 All this kind of applications will be trying to swing up. 58 00:03:29,000 --> 00:03:31,000 So there are many projects that are probably going to come up. 59 00:03:31,000 --> 00:03:35,000 But right now we will start with this basic project itself. 60 00:03:35,000 --> 00:03:35,000 Okay. 61 00:03:36,000 --> 00:03:41,000 The best idea about this particular project will be that, uh, first of all, we will go ahead in creating 62 00:03:41,000 --> 00:03:43,000 our entire project. 63 00:03:43,000 --> 00:03:48,000 Then second one is the will be setting up our environment variables. 64 00:03:49,000 --> 00:03:50,000 Okay. 65 00:03:50,000 --> 00:03:52,000 Environment variables. 66 00:03:52,000 --> 00:03:57,000 So when I say project, I will also go ahead and set up our own environment. 67 00:03:58,000 --> 00:04:00,000 Then environment variable. 68 00:04:00,000 --> 00:04:07,000 Then we'll see that what all libraries we specifically want like requirements dot txt will be implemented. 69 00:04:07,000 --> 00:04:11,000 Then we will go ahead and create our Streamlit web app. 70 00:04:13,000 --> 00:04:19,000 Inside the Stream Live app, I'll show you how to probably call the OpenAI API key the Lang chain. 71 00:04:19,000 --> 00:04:22,000 So these are like things that are happening in the industry. 72 00:04:22,000 --> 00:04:24,000 And this is what I will try to cover in this particular project. 73 00:04:24,000 --> 00:04:25,000 Right. 74 00:04:25,000 --> 00:04:31,000 And finally, after doing all this things, we will also make sure that we will do the deployment right. 75 00:04:31,000 --> 00:04:37,000 So all this steps will be specifically covered in this project. 76 00:04:37,000 --> 00:04:41,000 So just to give you an idea what exactly the project is all about. 77 00:04:41,000 --> 00:04:43,000 So here you can see this is what we are going to develop. 78 00:04:43,000 --> 00:04:47,000 So on the left hand side you can see you can put your opening API key. 79 00:04:47,000 --> 00:04:52,000 You can select different different models like GPT four turbo, GPT four uh or GPT four. 80 00:04:52,000 --> 00:04:58,000 So if I select this I can also set up the temperature value and the maximum tokens that I need to get. 81 00:04:58,000 --> 00:05:04,000 And let's say I will just go ahead and ask a question saying as high I'll be able to get some message. 82 00:05:04,000 --> 00:05:05,000 How can I assist you? 83 00:05:05,000 --> 00:05:10,000 Please explain what is generative AI. 84 00:05:10,000 --> 00:05:13,000 So I should be able to get the answer over here. 85 00:05:13,000 --> 00:05:19,000 Once I press enter, you'll be able to see that I will be able to get some kind of response. 86 00:05:19,000 --> 00:05:19,000 Okay. 87 00:05:19,000 --> 00:05:24,000 And, uh, based on the response, I will be able to see the entire data over here. 88 00:05:24,000 --> 00:05:27,000 So this is what it will get executed. 89 00:05:27,000 --> 00:05:33,000 And just to give you an idea over here, also, you can see, uh, the langsamt tracing is also happening. 90 00:05:33,000 --> 00:05:37,000 We will go ahead and deep dive on all these tracings and all also okay. 91 00:05:37,000 --> 00:05:40,000 So this will be taught in this particular project. 92 00:05:40,000 --> 00:05:43,000 So let's go ahead and implement this end to end. 93 00:05:43,000 --> 00:05:43,000 Thank you.