Everyone just give me only one minute actually it is taking a little time to load. Okay everyone we are live now so first we pack Talav Cry and after we start the session. Let me share my screen Aujbla min Shaitanjim Bismillahirrahmanirrahim Alhuz saj ful la taj ful fee amri wa tabahu fa Wakum tash Wakum fama wa maamin in feet qaumi tafkar Translation Allah Ta'ala is the one who smilingly created the sea for you so that the ships sail in it by His command and so that you may seek Maash by His grace so that you may be thankful and smilingly created for you whatever is in the heavens and whatever is on the earth, all of it by His command. Surely in this there are signs for those who ponder. Sadaqah Allahu a'zam Jazak Allah Okay everyone, let us start the session. Our main agenda today is different terminologies according to generative AI and we have Chat GBT. How does it work, what is its process, we have to understand this thing. So with us is Ma'am Doctor Taba and Miss Hasa hai to ji ma'am Hafsa over to you ma'am doctor doctor Taba over to you you can start welcome first let me share my screen. Bismillah Rahmahim Sir please tell me is my voice clear and the screen is being shared yes ma'am yes ma'am your voice is clear your screen is also shared ok so before starting the session a little introduction of mine I am an electrical engineer by profession and I have been serving at i guru for quite some time now. So what is our session about today? Today's session is about models of generative AI. Let us see today's agenda to see what we have to study today. Okay, so first of all, what are generative AI models? First we will look at these and how to find them. Ok? After that we will see what transformers are. Well, transformers are a general term for you, since I am an electrical engineer, for me it refers to the transformers etc. installed on the poles. These are the things that come to my mind whenever I hear the word transformer. But what is a transformer in generative AI ? That these are the brains of generative AI. Everything is processed from here and whatever input you give to any model, it gets processed from here and then generates the output and gives it to you. Just like CPU is the brain of any computer, which we call the brain of the computer, similarly there are transformers. Well, since mostly all of you here will be students, then who among you has used generator sorry chat GPT or any other such chat bot. You can re-hand it has been done by many people, 100 plus members. Ok. So a lot of people know how four GPTs and these kinds of models are working ah. Ok? So today we will see what you might have seen till now, what is its interface? If you give it any problem, it generates an answer and gives it to you and you can ask any type of topic to GPT or any AI model. Ok? Today we will see how he works. Ok? Other than that we will look at stemming and lemmatization. Well, these are some terms that you might find difficult right now. So you should not be worried after seeing this. We will discuss everything one by one. Ok? For stemming and levitation, you should understand that when you travel to any place, most of the people carry minimum luggage with them so that it becomes easier for them and their travel expenses are less. Similarly, when we use any Charge GP or any model, we try to explain our point to him in minimum text. Ok? So stemming and lemmatization is reducing that text. Ok? After that, let's go under the hood and understand the input requirements and how to measure the performance of these chatbots and generative AI models. After that we will see what API keys are. Ok? This is because we have to study it in this lecture and this exam is an important topic. So understand it now and whenever it is mentioned in the next session, understand its concept well. Ok? Apart from that, what is drag and its different techniques and tools? We will also see this at the last of this session. Okay, who among you has read about Artificial Intelligence or just heard the name about Machine Learning? Almost 200 people who are present here with us know what Artificial Intelligence and Machine Learning is or they have heard the name. Ok? This is enough for whoever you have heard the name till now. Good artificial intelligence is made up of two things. Artificial and Intelligence. Ok? We will see this now. One, it is a very large domain. One of its subsets is machine learning and one of the subsets of that machine learning is deep learning and then its subset is generative learning. So now let us see what all these things are? Well, we have a natural intelligence which Allah Taala has placed in our mind with the help of which we do all our daily work , read, write, do our work etc. Ok? Similarly, what the scientists did was that they developed artificial intelligence i.e. computers in such a way that they could think and take decisions not only about themselves but also about whatever data was provided to them. Ok ? So artificial intelligence is a type of intelligence that mimics human decision-making and the way humans think. Well, here I would like to give you an example that first you tell me that if there is a robot which is placing a particular type of object from one place to another, will it be intelligent or a robot which can lift any type of object and place it at any distance. He will be intelligent. First tell me whether the first robot will be intelligent or first yes first you tell me whether the first robot will be intelligent. How many people say that ? Ok? Almost 75% are saying that it will be the first one. Now those people should raise their hands who are saying that the second one will be a robot which can pick up and place all types of objects. Ok? OK I get the idea. Ok? The second robot that we have will be more intelligent because it will see every time that whatever object is kept with me like a book or a mouse or a pencil or a ball, what is its size, at what distance is it kept, and then it will move its arm robotic arm towards that object after seeing both these things, what is the size and what is the distance, how much should it move its arm and then to what size should it open the claw of its arm so that it can easily grab that object and lift it from one place and place it at another place. Ok? So this second robot will be called an intelligent robot. And because it will be robotically intelligent. There will be intelligence. So we will say that this is an artificial intelligent robot. Well, we saw in the previous slide that one of the subdomains of that artificial intelligence is machine learning. Ok? So what happens in machine learning is that we train the machine i.e. the computer in such a way that it acquires the capability of prediction and decision making by using different algorithms. Well, now if we look at it in broad spectrum, there are three types of machine learning. Supervised learning, unsupervised learning and reinforcement learning. First, let us see what happens in supervised learning. In supervised learning, whatever data we are inputting into the machine, we tell the machine in advance that I have just input some data. Let's suppose we input pictures of a cat or pictures of a dog to the model. Ok? So I will label it that brother, if I have all the pictures of cats then this is a cat and its color is blue, its color is brown or it is black or it is white. Similarly, if I give him pictures of a dog, I will tell him, what is the colour of this dog? What is the size? What is its breed? Ok? So and then I will train him and then when that model tells me that first I trained that model on all these pictures. Ok? Now I am asking him after training that module that I gave a new picture and did not label it whether it is a dog or a cat. Now I am asking him to tell me whether this is a dog or a cat. So now when it has to categorize an input into two types, then the output it generates will be called classification. Ok? Because it classified one input into two things. Ok? This and from this we can see that it is working on discrete data. It is answering us in yes or no. Ok ? This color can distinguish, like I gave the example of a cat, its color can distinguish and a classic example of this is spam emails, this email can be categorized into spam and non-spam folders. Well, apart from that there is another model we have regression model. It works on continuous data. Ok? And is predicting continuous values. Examples include housing prices, temperature, and weather forecasts for any region, or stock prices, all of which the regression model can predict. After that we come to unsupervised learning. What was in supervised learning? We were providing labeled data. In unsupervised learning, we train the model on unlabeled data and the model itself decides where to place the input. How he has to sort out the things he has. I will give you an example of this right now. Well, who among you has used Google Photos or do you know about them? Almost 200 students know about Google Photos. Ok? You can now Ray Ray down your hands. Well, I gave you the example of Google Photos because there is a feature in Google Photos in which what you can do is that it recognizes faces. Face recognizes facial features and on that basis groups similar faces in one place so that whenever it sees your face in this picture, it creates a separate folder for you containing only your photos. Ok? So what he's doing here is that he's grouping the photos separately, he's doing clustering here on the basis of facial features. Ok ? So this is how you can understand clustering. And the second is unsupervised learning, which is a model of dimensionality reduction. What we do in it is that we simplify all the data. We extract the extra information from it. That is, the model works in such a way that it deletes the extra information and stores only the critical information that remains. Now I gave you the example of Google Photos. Ok? Similarly, you can take the example of a photo, when we compress the photo, what will happen in the compressor is that the essential features of those photos or videos remain and everything else becomes invisible. Ok? After that we have reinforcement learning. Reinforcement means trial trial and error. Ok ? So what's happening in this is that you're modeling the model, the model is learning from your feedback. You used charge GPT here people. So you might have noticed that there is a feedback feature on four GPTs. You give him a thumbs up or thumbs down to his responses. On that basis, it generates the next response according to your preference. And similarly there is another feature on social media. When any social media platform suggests a new post to you on your feed, there is a feature in it that you can hide that post by marking it as not interested. Ok? So, you have given your input here and given your feedback that you are not interested in this post. You did n't like the topic. So what will he do now? What the algorithms being used at the back end will do is remember your feedback and will not suggest you similar posts next time. Ok? Let's move on. So what's next for Herity after that ? Deep learning arrives. Deep learning is a subset of machine learning. Ok? Well, what's happening in deep learning is that, first, I know how the brain works? Like you know about the neural network of the brain, there are neurons inside the brain and they are helping in decision making. Whatever your process of thinking and doing work is, it is all being controlled through the brain. And what cells are being used in that process? neurons that are in the brain in the brain. Ok? So what we're doing in deep learning is creating neural, artificial neurons. We create their networks and set their parameters in such a way that they can generate our desired output. Ok? So as the example is written here, the facial recognition features in laptops and phones are an example math of this. Well coming after that we have generative AI which is a subset of deep learning which is the last subset that we are studying right now. Now what we have done till now is that we have given some data, ah any kind of data and asked the model to make some prediction or do some decision making on its basis. Ok? But now what generative AI does is it generates new content. Which you might have neither seen before nor read anywhere. Ok? So what is its name? Generative AI. That means it is generating new content. It's not just analyzing the data. What can be included in this type of content? The best example is Chart GPT. It generates content for you in text form. Ok? Apart from that we have other ah generative models which do image and video generation. Ok? Similarly, we have a model called Cloud A which is being used for code assist coding and there are many more models which we will read about later which are doing finish moving. Okay, now let's come to LLM. Who knows about LLM here? anyone know? Ok. Ok? LNM is a short form. In this the first L means large. Second, L stands for language and M stands for model. Ok? That is, large language model. What is this? This is a model that has been trained on a large amount of text data, i.e., human language data. So it can identify all the complexities of human language. It can recognize and summarize human language text and generate new text of the same type. Ok? Well, the best example of LLM, which is older than four Gbit, is the autocomplete feature in phones and laptops. Ok? You can also understand that this is also a similar feature which LLM is already using. And it tells you by predicting what should come next after a word. Well LLM ah so far we've seen that human language can generate text like generate tax. Ok? Is. We have been using it for a long time. But how does this LLM work? So inside LNM because there is a model of deep learning. So what happens in deep learning? There are neural networks and using neural networks we create this and this is a and actually we read a word in our agenda, transformers, this a utilizes that transformer architecture and is processing the inputs through it. Ok? Well, what capabilities we have is that you can generate text from it. That means you can get essays written, stories written and code written. Ok? And this work of code i.e. generating code, we will also do it in this course. Many writers use it as an aid to writing their essays, stories, and novels. Good. Ah, apart from that, if you want to convert language translation from English to Urdu or any different language into your native language, then we also get the feature or capability of language translation from LNM and you must have seen that on some websites you have chatbot, that is, some websites are such like a food ordering website. Ok? So there is a feature where you can talk to the bot. Ok? So, the feature of talking to a bot that is coming there, you can ask questions about it there. That feature regarding that particular product has also been used there through LLM. Now we will see how it is used. Through API. OK charts GPT or Google Gemini and Floyd and Perflex these are some of the chart interfaces that we usually interact with in day to day life. Well, how do we interact with them, how do we get them to generate new content or how do we give our input, we can input text, we can input images. Ok? And if I have to present a lecturer, what I do is I get slides and text made from them. Ok? In this way you can use all these things. But what we have to see in this course is how we will use the API of these LLMs i.e. these LLMs in our solutions, whatever software solution we create. Ok? So we will use the API of these LLMs and similarly we have some specialized models about which we will learn later in the course. So we will see how we can use these in our software or websites. So let's move on to the next slide. Well this is the interface of Gemini. This is a four Gbit interface. Ok? So one of the concepts of this lecture is API. API stands for Application Programming Interface. Well, you don't have to worry about this word or its meaning right now because I will explain it to you right now. What comes in class one example of this is that if you go to a restaurant, what happens when you go to the restaurant? You go as a customer. There you call the waiter and get your order placed. I tell him to bring me this thing. The waiter goes to the kitchen and tells the chef that this order has to be prepared. The chef prepares the order and the waiter brings it back and delivers it to you. Ok ? So what is happening here is that when you as a customer are calling the waiter, what is the waiter you have here? API. Ok? You called the API to a waiter when you called the API. Ok? You told the waiter to bring me this or that dish from this particular menu as per your food order. So what did you tell that waiter what that meant? That you told that API to fetch this particular kind of data for me. Ok? What will he do now ? The waiter or in our case the API will go to the kitchen? The waiter will go to the kitchen. While where will the API go? near the server. Ok? And you can imagine the server as a kitchen here. In the kitchen, the chef must be preparing food. The server will process your data request. Ok? And then the server will bring it back to the API and the API will return your data request with the data that you have asked for. Just like you order food from the waiter and the waiter brings the food and delivers it to your table. Ok? This is a real life analysis. Now let us see how we can utilize it in our apps and solutions. Well, here I have taken the example of a weather forecast app, if you want to create a weather forecast app or website. Ok? What will happen there is that any user will come and let's suppose he asks for the weather of his city for that day. Ok? So whatever API you have used in your app there, it is most likely the API of the service of the Meteorological Department. Ok ? So that API will go to the server of the meteorological department. From there it will find out the data of your particular city. Ok? That server will find out that data and give it to the API and the API will fetch it and deliver it back to you. That means that data will be shown on your website that it is going to rain that day or the weather will be sunny or windy accordingly. Ok? Another example of this is to tell me the right calling apps like Ober, Yango or EndDrive. Which of you used these? So almost 160 people here have news those apps. Ok? Well, right there in the calling apps you must have seen that the map feature is used, that is, whenever you open that app, it asks you what is your location and where do you want to go. Ok? So then when you enter those two locations in it, then a map is shown on the map from your location to your destination. Ok? So the map used there is not such that every service, every right calling service uses its own separate map. This does not happen because mostly such companies are using GPS data. Ok? And by utilizing the GPS API, it became like Google Maps. By utilizing its API, it is using it on its sites so that it can provide the same experience to its users as Google Maps provides. Sir since it's been half an hour we can take questions. Shatla are you here? Yes ma'am, let us take some questions in the middle. So again we move on. OK sir. Okay I'll unmute when I call the name. Please kindly unmute your mic and ask the question. Yes Miss Ayesha Ayesha if you are speaking then I cannot hear your voice. Yes, no, actually I don't have a question. Initially, she asked a question regarding this. I think I raised my hand for that. I thought I can answer it. Okay ma'am, okay, let's take the next step. Adhnaata, have you unmuted Adhna aata? Okay, Javeria Zulfikar Abbas Khan. I think you guys are not able to unmute it. Assalamu Alaikum. Can you hear my voice? Yes, I can hear your voice. My question is what is the difference between an API and a Web API. Simple, can you answer it for me? Sir, in the question you said Simple API and Wet API. Web API, can you tell the spelling? Which API are you saying? Web Web Web or Web Web or ok Web API, okay, actually I was making a mistake in your spelling. Well, basically the main purpose of API is to make two different applications communicate with each other. Ok? Now what do you do inside the Web API? Makes the front end and back end communicate with each other. You are creating that API yourself. Right? Ok? That's you creating a path. Now this API, its main purpose is that the platform of different platforms will share a token or a key with you, which we will call API, what you will be able to do with its help is you will be able to assess multiple different models. Ok? The main thing is you are connecting to that platform using an API key to access something. Ok? Clear ji Assalamu Alaikum. Well, I just want to say that just like Tala Bhai explained about Web API and Simple API, similarly whoever is presenting the lecture or slides here should explain it in the same manner. At least, until the basics are not understood, then how can one read these slides? The slides are being read in the same way as is being followed in the universities of Pakistan. In my opinion, there will be no truth in this; all the 893 participants here will not benefit from it, so kindly, as Talb Bhai just explained it, in the same way it was explained. Thank you Sir, ok, I will call the names one by one. Kindly, whatever you say in the middle, don't say it. Ok. Ok Amman what mobile Amman sir mute everyone's mic sir. Then what happens after this? Salam Walekum Sir. Hello Assalamu Alaikum. Greetings to you too. Sir, I wanted to ask that in this course we will learn to train ML. Our main focus is to create applications using different models. We will not go towards Core ML. But we will definitely tell you the complete flow of how a model train actually works. Ok? We will do a couple of sessions on that. Ok. Sir, let me take my next question. Huzaifa Huzaifa Zaki Huzaifa Zaki Haftsa Jawad Sir, there was a question together, now Sajwad Sara Sir Sara, ok ma'am, let us move on a little further and then we will take the question again. Ok. OK sir. Salaam alaikum. Hi. Assalamu Alaikum Sir. I had a question that Exsa, if you are speaking then I think your mic is unmuted or muted. Ok sir. Yes, now let us see what problems can arise when we access this API and what are its advantages. Ok ? So let us first look at the advantages. Well, the first advantage is that you get increased efficiency and the speed of whatever application you have increases. That is because when you are using prebuilt functionality like I gave the example of Google maps and if there is an e-commerce website then it is using PayPal. PayPal is not available in Pakistan yet, but if it is using PayPal then what will happen is that it will not need to develop the functionality of that website like Google Maps from scratch. Ok? So the website can be created quickly because you are using the prebuilt functionality through API and your core focus is only on your business or your domain. Ok? Apart from that, there is cost effectiveness in this because it is difficult to make everything from scratch from zero. So when you are using any Prebit module in your site then you do not need to do any ground work for it. You use the system that is already in place. Ok ? So the cost gets reduced a lot. Ok? Apart from that, interoperability is achieved. How can you connect different software and different platforms? In that you do not need that everything you have is built on the same framework. Ok? Apart from that, you do not need to pay any separate cost for any other servers. Meaning there is no need to spend separately on them. Because when you are using the API of any company , using any platform, then the servers installed by that company will be processing your data for you also. Ok? And because GPT and Gemini etc. are very powerful models. Ok? So you have these powerful models that are trained on thousands and billions of amounts of data. And you should use their trend models in your site. So first of all, if you go to do data fact finding yourself, you cannot get so much data at once. It is very difficult. Ok? So to overcome this drawback, you use prebuilt models and those prebuilt models, since they are made for this purpose, are very powerful. Well, now let's look at its disadvantages that you have to keep your API secure on your site. Ok? So that you can avoid any malicious activity like hacking etc. Ok? So that you can secure your site. Apart from that, another drawback is that if you are using the API of the company or the site like GPT or Gemini or AWS Amazon Web Services, then if it ever happens that the site goes down or there is some malfunction in its server, then due to that your site also gets disconnected and that functionality gets reduced or discontinued on your site, your web website or your application. Ok ? So if there is a downtime on the API of the company that owns it, then your site also gets affected and business operations get affected. Ok? Apart from that, because there is communication from server to server, your site may become a little slow due to this and internet connection is required for this. Ok? Well, till now we had not read about models. So now let's see the models that we have and which models we will be using most of the time. We have speech to text generation models. That means you dictate in this. Whatever you want to say, it gets converted into text. Ok? You might have experienced this feature even on 4 Gbit when you are dictating to your processors. Apart from that, there are text to text translation models of Google. There are object detection models. Image generation is text to image conversion models in which you process text. A And in that text you tell me what kind of image I want. For example, when Mid Journey went viral some time ago, its feature was that it converts whatever you say, i.e. text, input in your text form, into an image. Ok? Apart from that, there is also a model to classify faces through edges. Face Fase edge detection model. Well, till now we have seen what models are and how we can access them through API. Now where will we get these models from? Ok? So we have a site. Well, there are other sites too. But mostly we will be using Hugging Face in this course. Ok? Let me show you its interface. Well here is its interface. Here you can see that its models, database, data sets, all these are coming here. Ok ? So, let us look at its models first. Look here is the text to speech generation model of Mr AI and here you will find many models. Apart from this, if you need data sets then you can get them from here and since we are running a course, we will try to use free models. So you should also log in to your account by going to Hugging Face so that you do not face any problem again. Well, it was all from my side. Now the next slide deck and the remaining topics will be conducted by our second trainer. So over to you ma'am Tayba. But if Sartala wants to take some questions then I can answer them. Okay, we'll just take some questions for two to three minutes. So then ma'am hands over to Taba. I'll unmute one by one. Kindly do not unmute the center mic. I am saying it again. Yes Sajwad Habsa Jawad Abbas Khan Assalamu Alaikum Sir I have a question Sir this ma'am who is teaching will also show us its practical and if these slides are shared in the group then it will be very good because those who have joined late and have missed it, they can read it so that it will be easy for them. Okay, let me just add this thing. All these slide recordings etc. are shared with you. If you go to the group description and see the WhatsApp group in which you are added, then there is a link attached to it of Google Document or Google Sheet, so if you click on it, you will get these slides from there and secondly, as far as practical is concerned, we will do everything in practical. Today 's session was an introduction to terminologies, in which we are discussing the different terminologies that you might be using during the course. Ok ? So today's session is the main session and Tuesday, Wednesday, Thursday will be practice sessions based on today's session. Ok? clear? All right sir. All right. Ok. So let's take the next question. This is my question. Okay sir, you can kindly repeat what you told me to log in first. The login information that the lady was telling me about can it be repeated by logging in yourself. Kindly sir, I will do it. Sir this is your site Hugging Face and here because my account is logged in. So if you search on Google, you will get Hugging Face. Let me search it for you. Hugging a minute. Sir, I opened a new tab, can you see it ? Yes ma'am your tab is showing. Okay, so here you search, there is a site called Hugging Face. Ok? So here you will go because my account is logged in, so because of this you can already see everything here. But when you go here, I'm trying to log out and show it. Let me check that I can. Sir, please do ok, I will show it to you on the main page. Ma'am, if you please stop the screen. Ok sir, I will show it to them. Okay let me know my screen is being shared with you please tell me this confirm that the screen is being shared ok what do you have to do here just click on hugging face dot I have attached its link with you ok I have shared this link with you inside the chat and similarly this link is also added in Google Doc Google Sheet. So you will see here on the right most side you will see login and sign up. Login will be used by those who have already used it. So if they log in, they will be logged in. Otherwise you have to come to sign up. After signing up, it is similar to the steps you take to create your social media accounts. You have to enter any email address that you use and keep the password. Have to do next. So a complete video about this, which is CC, how will you log in, that will also be attached. Ok? That too Google sir, I will log in. Let me tell you a little about Hugging Face, what is it and what is it? Basically about Hugging Face, Hugging Face Sir, it is the Jumma Bazaar of models. Ok? In simple words, all the first time models which are free of cost and are open source, you will find them here. Ok? Simply you come here to the models. Ok? These came on the model's tab. We talked to him on chat GPT. Chat GPT is used by all of you. Now the Chat GPT that is running nowadays is 5 5.2. Obviously, Chat GPT one came, two came, three came and so on, these updates would have happened further. Ok? So here we search Chat GPT. Which models of Chat GPT are available here which are free? Ok ? So here these people have done something or the other. Ok? This specific name has been found and tuned by people. If you want to do Open AI, because Open AI is a company, then you will do it on Open AI, so look, there are multiple different free of cost models of Open AI. GP2 is lying open sir, what does it mean, yes sir, kindly tell me what is its use, what is its real purpose, for what work can we use it, we have to pick models from inside the King Face website, I am not able to understand it, Salam Walekum Rahmatullah, I feel that many of the participants do not know the basics of what is open AI and this, okay kindly if you take some help from me, I think so I am trying to answer it, I am telling you people, right now it is only a session on different terminologies. We will use all these things. There wo n't be anything that will just pass you by. Ok? Overall, we have a total of five weeks in which each thing is divided into sessions. Okay, sir, there is a point in using the chat, I want to tell you this only that when we do practical in future then I will teach you this thing, yes, so that's it, right sir, I had a question for you that nowadays in LLM, that Large Language Model, as it is the era of LLM. Okay, can an individual developer fine tune these on his local system or is this only the work of big companies? You can fine tune but building it from scratch is not possible. Sorry sir, I had a question on LLM that means we can use it on cloud, right? We cannot use it on a normal laptop on a local machine. If we fine tune it, then for that we will need cloud or super computer. No sir, sir I have a question, please move further. Why are you asking questions about all those who have not read it? Sir, my request to you is to keep these questions open only for the host. You will get questions in text, you can answer them. Instead of stopping for a while in between, I think we will keep only the Friday session on question answer and rest we will take you people in the same typical way. Ok. I don't think that any person sitting here is from sixth or seventh class. All of them are either university graduates or university students. And they are mature enough, everyone must understand the matter in a simple way. Kindly make sure we will do all these things. Ok? I am sharing the document with you all once so that you all can check at your end when this thing will be shared. Instead of repeating the question again and again the same thing we have to do. This sheet was shared with you yesterday also. Attached inside the WhatsApp group. It is written in the group description. From there you can assess it. If you come here, this is our main document in which the complete outline is shared which we have to do in these complete six weeks. Today if we talk about what we have here, there were two main sessions. Ok? There are 8 to 10 day practice sessions and a hackathon.
HEC Generative AI Training Program | C3 | Week 1 | Main Session 1 | Monday | Part-1 Topic: Introduction to Generative AI