FULL STACK DATASCIENCE BY VIJAY SIR DURGASOFT YOUTUBE LIVE STREAM

追加: 2026-05-07

[00:00:16]Yes. Yeah. Good evening all. So myself I'm Vijay like I hold 20 plus years of experience in different technologies. So before I start can anyone please confirm whether my voice is clear whether the screen is visible or not. Right. Yes.

[00:00:35]>> Anyone?

[00:00:38]>> Yes sir you are audible.

[00:00:39]screen is also visible.

[00:00:41]>> Fine. Thank you. Uh fine.

[00:00:45]So data science full stack data science and AI artificial intelligence right. So this course totally deals with the entire data science machine learning concepts deep learning total deep learning techniques and AI and advanced AI right advanced AI. It's a totally 5 months course. So I'll be totally briefing you about what we are going to discuss right and what are the advanced things advanced topics in AI especially fine so just I'll give I'll just give an overview about this data science and AI what is agentic AI generative AI so first thing is data everything deals with the data data has got experience data has got future reference using data and using data only we perform data analysis, data predictions, data classifications, clusterings, right? So the entire world, the entire world flooded with user data.

[00:01:46]The entire world is flooded with user data. Using this data and using this data science, we perform this various operations. So here data science data science combines it combines maths statistics.

[00:02:06]It combines both maths and statistics and apart from that also like programming programming like Python programming and advanced analytics.

[00:02:19]Advanced Analytics, AI, artificial intelligence that is AI, machine learning for extracting insights. for extracting insights.

[00:03:03]Organizations organizations data.

[00:03:07]Understand here data science combines maths, statistics, programming, advanced analytics, artificial intelligence, machine learning for extracting insights hidden in organizations data. in a simple way, in a shorter version. If I talk like in a shorter version, data science just I'll keep in the data science is the art of turning it's art of turning data into actions.

[00:03:44]Data into actions.

[00:03:48]We talk like a data science focus on value.

[00:04:00]It focus on value not the value.

[00:04:05]It focus on data that is focus on value not the value.

[00:04:21]So data science is a field that is evolving at a very rapid speed today to be so be part of this journey and data science emerged from statistics and data mining. So data science is all about using tools, techniques and creativity to uncover the insights hidden within the data. It combines as I said it combines math, computer science and the domain expertise right to tackle with real world challenges in variety of fields. So in simple way data science going to process the raw data solve the business problems even make predictions about the future trends or the requirements.

[00:04:59]For example, you got some huge raw data in a company. Data sense can help you answering various questions. What do the customer want? How can we improve our services and what will be what is the upcoming trends in the sales? How much stock is required for a particular event. So all these things right. So it is going to help and people generally know talking about this AI and if you see AI generative AI in a simple in a simple way I'll just describe generative AI so ju just one second just a second now once again anyone can you please confirm whether my voice is clear.

[00:05:54]Yeah, once again can anyone please confirm whether my voice is clear?

[00:06:00]>> Yes sir, it is clear.

[00:06:01]>> Fine. Thank you. Just generative AI creates new content creates new content like text like text or images. It create text or images by learning patterns from the data.

[00:06:22]It learns creates new content like text or images by learning patterns from the data.

[00:06:29]I talk like agentic AI.

[00:06:32]Agentic AI or if I say AI agents and making decisions and making decisions and executing and executing multi-step actions.

[00:06:57]See, it's decision making and executing multi-step actions.

[00:07:10]to achieve the goal in a simple way.

[00:07:17]Generative AI generative AI is about creation is about the creation.

[00:07:26]Agentic AI is about action and decision making.

[00:07:38]And decision making. Just understand this. Generative A is about creation.

[00:07:43]Agentic A is about action and decision making.

[00:07:52]Okay.

[00:07:55]Agentic generative AI. See, see this artificial intelligence, artificial intelligence, right? Developing systems that can perform the task. Okay, I'm increasing my voice. Some of you are not audible, it seems, right? So, developing the systems that can perform task that typically require human intelligence such as learning, problem solving and decision making.

[00:08:21]Right? Yes, we can say generative AI is the subset I can say generative AI is the subset of AI designed to create a new original content new original content it learns patterns structures from the waste data sets that is the vast data set and then uses the knowledge to generate new outputs to generate new outputs like text images or code example like charg for text L E for images and various tools for code generation that we'll be discussing right. Yes.

[00:09:02]>> Is my voice clear to all? Can anyone please confirm me once again?

[00:09:08]>> Yes.

[00:09:09]>> Yes. That one one or two are keep on responding that Suni Lawasti is my voice not audible.

[00:09:18]So everyone know okay so everyone has no issue right okay I think it's from your side right okay so everyone has got fine it seems okay fine so I was talking like examples like charg for the text and dal e for images and various tools for code generation, purpose, content generation, enhancing the creativity and if I talk like agentic AI, agentic a system designed to make decisions, take autonomous actions to achieve a specific goal.

[00:10:09]Okay.

[00:10:11]Now see uh so most of them they're asking so just what I'm going to discuss as part of the course I said it is like total like full stack data science with advanced AI so what are the things I'll be discussing step by step I'll be discussing what is the part of our course if I give course overview one Get that.

[00:10:49]Okay.

[00:10:55]Python introduction.

[00:11:00]So uh we are using a Python for the implementation right.

[00:11:07]Python is the mostly chosen language for implementing this MLDDL all these things.

[00:11:12]It has got many flexible features. It has got many modules for especially for data science environment like numpy pondas mattplot lip cpskit and tensor.

[00:11:22]In this way many modules python is providing compared with other programming languages.

[00:11:28]and later like data analysis data analysis using numpy and pandas.

[00:11:37]So data analysis using numpy and pandas where we'll be discussing about uh data frames in pandas numpy arrays array 2D 3D dimensional arrays and numpy flattening reshaping working with this pandas data frames how to create this data frames from different objects working with different aggregations merging can different operations using pandas data frames and also like data visualization data visualiz ization right using plot lily and math plot liplot statistics for data science statistics for data science see generally statistics helps a data scientist understand the data variability the testing the hypothesis summarizing the complex data sets We have got different like descriptive statistics, inferial statistics, dispersion measures, sampling techniques. In this way, we'll be discussing on this statistical methods, [clears throat] data distributions, types of sampling, confidence level, correl correlation analysis, use of correlation, continuous variables, categorical variables. Right.

[00:13:00]Next like uh machine learning techniques machine learning techniques so here we'll be discussing about the different types of ML algorithms linear regression nonlinear logistics decision trees random forest classifications clustering techniques [snorts] partitioning methods hierarchal clusterings right FB growth in this Yes, time series analysis, natural language processing, NLP MLOps machine learning operations set of practices that combine a machine learning development and the operation devops to automate streamline entire ML life cycle.

[00:14:19]Okay.

[00:14:23]And also you'll be getting knowledge on this uh like big data spark that is spy spark we say very high speed processing technology right it's a execution model with very high speed you can process the data like why pi spark need for it and pispark benefits to the professionals the execution architecture those things right so NLP and text mining See uh this NLP it allows computers to understand, interpret and generate human language. It is going to bridge the gap between the people and the technology.

[00:15:09]Here we use NLTK natural language toolkit. It is a suit of libraries and programs for symbolic and statistical natural language processing written in Python programming language deep learning.

[00:15:36]Okay.

[00:15:40]In this again an annn CNN artificial neural networks convolution neural networks be getting tensorflow.

[00:16:00]So we will be discussing more on this deep learning techniques. TensorFlow it's a free open-source software library right for machine learning and deep learning. It is developed by Google. It provides a comprehensive set of tools, APIs and libraries to help the developers build, train, deploy machine learning models efficiently across various platforms.

[00:16:25]Yes. Next.

[00:16:38]One second again some clear for me. So consider duration of the post is monitor. So yes for your general queries right? Yes Ganesh right for general queries I'll be giving time for you at the end of the session. I'll be briefing you about all those things. Right. Yes.

[00:16:57]Convolutional network. What is convolution? what is a convolution layer network and filtering pooling data flattening fully connected layer all these things I'll be discussing and uh uh more on this deep learning techniques BSMAN So Boltman machine and autoenccoder it's a type of neural network that uses probabilistic approach to model data distributions unsupervised learning model with visible input and hidden nodes right and also like autoenccoder autoenccoder I'm queries one second transform yes yes yes it will be covered right yes nicl autoenccoder it's a type of unsupervised deep learning al that learns to compress data into lower dimensional representations and reconstructing back to its original form and next like we're discussing more on this G A N generative adversarial network GN generative adversarial network in deep learning. This GAN is a powerful model architecture that uses competi competing neural networks. So like a generator discriminator creating some fake data samples while the discriminator role is to differentiate the fake samples and then later I'll be discussing about RNN gru RNN recurrent neural network it is a type of neural network specifically designed to process sequential data and that to GRU gated recurrent unit it's a type of recurrent that is a neural network designed to process sequential data and LSTM long short-term memory it's a specialized type of RNN that is a recurrent neural network in deep learning to process sequential data by selectively remembering the forgotten information over the long periods Yes.

[00:19:32]Yes.

[00:19:36]>> Open CV.

[00:19:39]>> What is sequential data?

[00:19:48]>> Yes.

[00:19:50]I'm not just I doesn't want to enter into the technicals as of now. just what we are going to discuss Open CV it plays a vital role in deep learning open CV like it's a deep learning particularly within the domain of computer vision computer vision like it serves a powerful toolkit this open CV it provides a powerful toolkit for handling the image and video data that keeps deep learning model processes and uh coming to this this topics right next I'm coming to this AI generative AI gener it's a subset of AI just before I said right creating a new content like whatever the text or images right learning patterns from the data learning patterns from the data right yes prompt engineering Someone uh keep on messaging right facing problem with the voice. One second. Give me a second. Give me a second.

[00:22:59]Hope now it is fine to all. Once again can you please confirm?

[00:23:06]>> Yeah, once again can anyone please confirm right? Is my voice clear to all now? It will be >> sir. Clear.

[00:23:16]>> Yes sir.

[00:23:17]>> That's fine. That's fine. Okay.

[00:23:18]Generative AI, prompt engineering, prompt engineering, designing and optimizing prompts for AI models, especially large language models, LLMs to accurate and and to get the desired outputs, right? Yes. And also like advanced prompt techniques, advanced prompt techniques and LLM APIs. I'll be discussing working with LLM APIs.

[00:23:46]Okay.

[00:23:49]and also building this LLM APIs with longchain and LLM index. Right? Yes.

[00:23:56]Developing rag systems or a rag systems.

[00:24:06]RAG is nothing but like a kind of retrieval augmented generation systems AI framework that enhances LLM's large language models combining the external information retrieval systems and also like uh vector databases vector databases and embeddings vector databases and embeddings right yes vector database stores like unstructured data such as images, audio and highdimensional numerical representations what we call as vector embeddings.

[00:24:49]Fine.

[00:24:55]Yes. Yes. Yes. Yeah. Very aspect. Yes.

[00:24:59]Ganesh. Yes. It is covered.

[00:25:08]Next object large language model operations specialized set of practices and tools designed to manage uh what we say the life cycle of large language models even in the production environments.

[00:25:28]They are like subsets of ML ops right?

[00:25:31]Yes. Next I'll be discussing about agentic I said right system to make some kind of decisions and taking some actions to achieve your specific goal. Right.

[00:25:49]Right.

[00:25:52]And also different other you can see the various other topics within this agentic. Right. the architectures and design patterns and also like working with the longchain building AI agents building that and AI agents with long graph implementing agentic rack and multi- aent systems and AI agent observability and many more right you can just check the course content you can get a brief knowledge on it and also like even I'll be showing you on this cloud this uh ML, cloud ML, AWS, Sage Maker and Google Cloud AI, Azure ML and the serverless ML right yes these things and this deployment process model deployment okay once just have a look on this course content right you will get the total knowledge so starting with this data science introduction why data scientists are in demand today growing need for data science data science use cases is data formats, data quantity, data quality, data transformations and coming to the Python for data science, Python functions, [snorts] collections, list sets, dictionaries, numpy for numerical, Python for data analysis, numerical data, pandas for data analysis, math plot lip, Python for data visualizations, statistical methods for data science, types of statistics, Central tendency measures, mean, mode, median, the story of average, data distribution and machine learning techniques. Everything in brief, time series analysis and forecasting, MLOps fundamentals, [snorts] big data like big data, pi spark, right?

[00:27:44]very high speeded processing execution model NLP and text mining overview way of it extracting cleaning and pre-processing text deep learning tensorflow convolution neural network boltsman machine and autoenccoder generative adversial network emotion and general detection introduction to RNN and GRU LSTM open CV generative and see up to module 20 onwards. See uh we'll be going deep brief into this um AI advanced AI from module number 20 to 40 we'll be discussing a lot on this prompt engineering advanced prompt techniques coot to right change of thought tree of thought working with LLM APIs building LLM apps with longchain right developing rag systems vector databases and embeddings building end to end generative AI applications s evaluating genai applications in enterprise use cases multimodel LLMs LLM ops and evaluation agentic AI see this model again from 31 to 40 I'll be discussing on this agentic AI agentic architectures and design patterns working with longchain AI agents and long topics implementing agentic rack developing AI agents with feed data uh multi- aent systems advanced agent development with autogen building a agents with no low code tools right so in brief in depth we'll be discussing about this as compared with other training sites right you can see we can compare this course content right so advanced AI topics we'll be discussing that's why I said it is uh for 5 months it will be for 5 months in brief 3 months months for AI two months we'll be discussing for data science >> you'll be covering oops also >> no no no see uh you'll be given the videos for that if you want in brief in depth on this you can just go through the videos of this Python but whatever the basic Python knowledge required that will be provided because the course duration will be increased if you keep discussing all those concepts but I'll do overview about all those things.

[00:30:18]>> So is it mandatory to know everything like oops Python?

[00:30:23]>> No no no we are not developing any web applications here >> not required.

[00:30:28]So up to what like uh modules how to work with the modules how to work with functions how to work with collections how to use that variables constants these things are enough Python collections you need to know Python functions you need to know Python modules these things right and uh the offerings from my side each and every session recording live session recording You'll be getting the live session recordings even if you miss any session or couple of sessions right you can check with the recording and attend the next session for the better understandings.

[00:31:11]So each and every when the session completes each and every day every day one video will be added to your Google drive.

[00:31:19]Clear soft copy of notes.

[00:31:22]Clear soft copy of these notes will be provided topic wise assignments and tasks assignments and tasks to work with and uh so generally I don't know like about other trainers other in I want to have direct interaction with the students so that's why I'll be creating a separate WhatsApp group so directly where you can interact with me or if you got any errors You can take snap of it and post in the WhatsApp group. Anyone can respond. I'll be responding right the separate WhatsApp group for the batch assignments sorry uh >> assignments will be given daily or weekly >> interview. So topic wise I'll be providing with interview questions interview questions around 200 interview questions mostly asked questions live interactive hands on sessions even some things where you want to do par along with me you'll be doing that installations or other things.

[00:32:43]Apart from this, right? Yes.

[00:32:48]Apart from this, uh this is from my side. Live session recordings of copy of notes, assignments, tasks, WhatsApp group, interview questions, live interactive, hands-on sessions. Plus, uh you'll be getting the videos. I said all these videos you'll be getting means what we'll be discussing daily basis.

[00:33:03]You'll be getting the videos plus uh you'll be getting the Python videos Python videos around if you want to have in in brief in-depth knowledge on this 100 plus videos on this Python right 100 plus videos on this more than 100 videos you'll be getting on this Python you wherever okay you can just whenever you you can just go through this for better command on this Python as just now know someone asked and and apart from this like even MySQL videos for database related these things will be provided Python and MySQL videos will be provided this will be free free access for one year you'll be having one free access for this videos this will be provided by Dulasoft and this is the offerings from my side and all the sessions will be interactive sessions at any point you got any query you can immediately unmute and you can ask your queries.

[00:34:05]Yes, >> sir. Uh is it possible to add fast API or streaml library as well because we may have to um once the model is developed we can create the APIs or uh basic web app.

[00:34:22]Yes. Yes.

[00:34:25]Yes. So whenever required wherever required means u I'll pre-intimate if there are any extra sessions I want to take because the syllabus is to you can see like we got lot of things to be covered so some sessions on the Saturday so classes will be from Monday to Friday but sometimes on Saturdays or some other sessions whenever I plan so it will be pre-intimated for you everyone fine with the timings we'll be going with the sessions fine everything uh even people who are from non-programming everyone right you'll be getting the prerequisites right even I was saying like some big data like pispark which has got a great boom that I'll be discussing and uh in brief for AI we require some deep learning techniques like more knowledge on this deep learning that everything I have discussed there and also like 3 months around 3 months around I'll be spending on AI and 2 months for this machine learning deep learning learning and all these things right so >> learning these concepts can we build a application like GPT like will you teach us a project >> so we'll be yeah we'll be I'll be discussing like three projects in this course so we'll be developing those okay so yeah so this is something that is what we are going to discuss as part of this so that's why it is full stack data science and AI Right advanced AI means in brief in depth we'll be discussing about the AI topics that two people generally asking for generative A and aentic A in that you can see that what I am going to discuss in brief about this generative AI and aentic AI concepts okay >> like what kind of jobs we'll be getting after learning this >> so AI engineer so you can go as many many jobs like a you can go as AI engineer machine learning engineer uh data scientist you can go so many roles even you can see today if you talk about this jobs right uh this data sense in AI the top most jobs you can see in the current from 2000 if you see from last 2 three years right so top highest paying jobs these are the highest paying jobs you can see so I say just you as you asked like we have got many many uh what we say career kids careers in this data science data scientist and even if you add some other some few things to this you can also go for data analyst also but you need to have some knowledge on this PowerBI and the other machine learning engineer right and uh business analyst you can go but some few things you need to add for those things and uh coming to this uh if you say if you see this package case also you can see great packages for this AI. If you got knowledge on this AI uh if you see data science jobs are the top most ranking jobs if you see uh many companies if you talk about many companies like Facebook, Microsoft, Amazon they are interesting in hiring data science and AI engineers with great salary packages and also like tech gens like many visit the companies like Google, Apple, Cognizant, Wall mode all these they are offering high paying jobs to the data scientists I said these are the top most paying jobs and with the highest salaries data scientist has got the highest salary packages so pursuing this data science career right data science in this AI so it's the best choice right it's the right time to just go with this data science is in high demand growth of digital data even combined with advanced technologies like AI and machine learning. So if you are lot of so the industry is in required there is a lot of requirement for this professionals who has got great skills.

[00:38:35]So they'll be providing you like a huge bigger packages in the same way you need to showcase your skills right. So you need to have some you should have strong technical skills based on this. So that's the reason right we'll be people who has got the knowledge and a that with advanced topics advanced techniques right.

[00:38:54]So solving many real world problems training some advanced systems like just like human intervention.

[00:39:04]So that is that's what I want to say like it's most promising and high in demand career path people who want to choose even the beginners and even the freshers also right if you have got this much of knowledge and if you showcase your knowledge when you get a interview chance definitely you'll be the first choice.

[00:39:22]Yeah. So if we learn this knowledge this will this will be enough for data science roles, scientist roles.

[00:39:30]>> So even lot previously like uh mean some of my students they got jobs as a fresher itself.

[00:39:40]>> One month ago right?

[00:39:46]>> Okay.

[00:39:46]>> Yes. Class timing. Let me let me make it clear about the how data engineer difference with this. Okay. So some query was asked how data engineer different from a okay a IML right yes see data engineer right okay data engineer also one of the pre promising we can say today today in today's industry right both data scientist and data engineers have what good roles are great roles right they have got good salary packages data engineer is for data management simply if I can say data engineers are for data management data storage data processing data filtering, data merging, all these they have to play with the data, right?

[00:40:27]Data engineers, right? U they are going to extract the data, they are going to process the data, they are going to filter the data, merging the data, grouping different based on the business logic, they are going to uh transform the data as per the requirement.

[00:40:44]They perform data management, data storage and data processing. Here in this case how it is different from this ML and this means here we perform not data management like data analysis data predictions classifications recommendations all this we perform here as compared with data engineer data engineer for data management practice sir how many projects will we cover for ML and >> so around three projects I'll be discussing Okay.

[00:41:18]>> And three projects will be parallel or after completion of everything.

[00:41:25]>> So after see after completing of data science one I'll be discussing after completing of this AI I'll be discussing too.

[00:41:36]>> Okay.

[00:41:37]>> And uh let me make it clear about the timings as it is uh too late some of them many requests were made. So uh just one second.

[00:41:49]timings first week it will be at 10 p.m. it will be first one week it will be later after one I'm making it clear about the timings right before you go with the payment process it will be at 9:00 p.m. First one week only it will be starting at 10 p.m.

[00:42:12]later after one week one week to it will be at 9:00 p.m.

[00:42:18]Okay.

[00:42:19]>> Is it daily 1 hour or 1 and a half hour?

[00:42:21]So initially initially I'll be taking for 1 hour means once I start with machine learning concepts those things I'll be extending it to one and a half hour depending upon that requirement right so first initially it will be for 1 hour but later once we go with the technical things right when I start with machine learning deep learning I'll end I'll just extend it to one and a half hour okay so wherever required I said I'll be taking extra sessions I'll pre-intimate you about those things >> and uh which module you will teach and which module you will share the videos which module okay that you are saying right no maximum I'll just touch each and everything each and every each and everything I'll be discussing because people there are people who are from uh basic doesn't have an so keeping all the students in mind we'll be discussing that's why so wherever require some topics I'll be taking extra extra sessions for some of the topic as you said which models which we won't discuss not like that but uh I'll be discussing but I'll take extra sessions for that >> for example python and numpy pandas matt for this one it will take around 3 months right >> no no see actually it is not a python course right yes once once that's what I'm saying it's not a python course >> uh total python will be discussed total python programming two and a half to three months everything not like that.

[00:43:51]So we'll be spending more time for our core concepts means like data science, machine learning, deep learning and this AI we'll be spending more time right even for with Python, Pandas and NumPy like for all these things I'll be sparing like two weeks time.

[00:44:09]>> Okay, >> I'll be allocating two weeks time for this uh 15 days time for these things.

[00:44:18]So later later right you can check with this videos right Python videos you can just go through if you want to have full content on this. So basic whatever I'll be discussing the 15 days it's more enough right for performing the things Python numpy pandas visualizations.

[00:44:37]Yes. Yeah. Someone asking a query right?

[00:44:39]>> Yeah.

[00:44:40]>> So till module number 20 um all the modules will be covered in the first two months only and then from generative AI it will take 3 months. So all these um modules till module 20 are these prerequisites for learning generative AI or they are just uh additional um knowledge based for this uh whole course.

[00:45:01]>> Generally you need to have some knowledge on this right this topic from model number 20 onwards. See uh there will be some prerequisites basic knowledge you need to have on this uh the basic statistics knowledge should be there some ML knowledge should be there right see generally one thing I'll say one thing I'll say your name what good name >> uh may >> right if you if you just hear frankly speaking here generative words 20 model number 20 to 40 generally generally people who are discussing only AI they'll keep AI course as artificial intelligence >> okay >> see if you see other training institutes or other trainers right who keep only AI artificial intelligence they'll keep the heading the course name is artificial intelligence but if you see that artificial intelligence maximum their syllabus off of 60% of the syllabus they'll be discussing about machine learning deep learning only that come to that generative agent a only like last one page will be Yeah, if you just compare and check, right?

[00:46:09]>> Mhm.

[00:46:09]>> If you compare, they'll keep the heading as AI only. But mostly again they'll keep Python. Uh again they'll keep statistics, they'll keep data science, machine learning. Even 60 to 70% of that AI will be these things and only 30% they'll discuss about this generative AI and agent. You can check with that right.

[00:46:30]>> Okay. Yeah, >> you can check with that. But here not like that we are teaching separate separately means everything total knowledge you'll be getting about all these things. Yes even uh students also where you'll be getting the entire knowledge about all these things right.

[00:46:44]>> Okay. And yeah one more question sir. So for this um uh vector database and embedding. So yeah we'll be um learning from scr uh scratch to set up this vector DB and uh um uh configuring the indexing part.

[00:46:59]>> Yes. Yes. from that everything from the basic ground level I'll be discussing.

[00:47:03]>> Okay. Okay. Thanks.

[00:47:07]>> So okay for those things right somewhere you two need to support right by taking [clears throat] some extra sessions right because uh is time taking and sometimes it may go to 2 hours also 9 to 11 also in some of the days because I cannot stop in the middle of that. So wherever required I'll say 1 and a half hours.

[00:47:28]Sometimes it may also go for 2 hours.

[00:47:31]>> Uh there's another batch in morning also, right?

[00:47:35]>> Yes. Yes. So most of them also asking for the morning batch. So there is a morning batch also given for that.

[00:47:42]>> So it's a 8 or 9 a.m.

[00:47:44]>> 8:00 a.m.

[00:47:46]Okay.

[00:47:48]>> So for this one, uh it will always be the night classes, right?

[00:47:53]>> Yeah. This will be the same thing.

[00:47:54]That's why I'm making it clear about the timing. So people so it will be 9:00 p.m. it will be starting first one week to 10 days it will be at 10 p.m. but later it will be starting at 9:00 p.m.

[00:48:04]always that classes will be in the night evening only I don't disturb the timings even I make sure that each and every session is important to the student even I want the students to interact in the class I'll keep that entire uh I'll unmute it will be unmuted status anyone can ask that you should keep on responding for the queries I ask it should be a live interactive sessions from both the side >> not like not like other trainers like they'll be keeping in unmuted status at the last 10 15 minutes they will ask you if they what got any queries not like that anytime you can ask queries anytime you can pass and uh >> yeah that's um sir is it uh MCP included in the agent part >> yes yes >> cool thanks >> and uh here separate WhatsApp group I'm maintaining for the students to have direct interaction with me if any errors if anything you can see this Okay.

[00:49:05]>> Can we keep three products in our link in like what kind of projects we discuss >> okay fine I'll I I'll int any other queries you got >> so morning time uh 8 8:00 a.m. right?

[00:49:27]>> Yes. Yes. And that is also mean starting from uh today or uh tomorrow.

[00:49:33]>> It's already start.

[00:49:35]>> No no it's uh from next Tuesday it is >> okay next Tuesday.

[00:49:41]>> So so actually previously there were data science Can anyone please confirm whether my voice is audible now?

[00:51:27]>> Yes.

[00:51:28]>> Yes.

[00:51:29]>> Yes.

[00:51:30]Right. Okay. Someone was asking a query the last everything from the basic. So just I was saying that previously uh they were like uh 1 second.

[00:51:56]>> Hello.

[00:51:58]>> One second. So previously data science it's a separate course. We were taking data science separately.

[00:52:12]artificial AI separately, right? There were two courses separately.

[00:52:17]Uh it was for 20K. This was AI for 20K. But it is an offer batch where combinely both AI and data science and AI just for 20K it is provided. So this is for the first couple of first this and next this for this first two or three batches right means data science and advanced AI in this advanced AI. So where we have combined both data science and I combined both in advanced a advanced AI topics. So it's just offer batch it's 20,000 plus you'll be provided that recordings of nearly like 8,000 free recording access on Python more than like 150 videos on Python and like around 35 to 40 videos on my SQL right who doesn't have data west knowledge right yes so the timing is this 10 p.m.

[00:53:14]and 9:00 p.m. So tomorrow also everyone can attend the session same time at 10:00 using the same link to see more discussion right yes okay so if I'm done with that queries if there are no other queries I'm signing off for now meeting tomorrow same time >> sir >> do you have any weekend b starting soon >> weekend batch not in this next one month no see one thing there is a weekend batch going to start in November month.

[00:53:47]But one thing people if you got enrolled in my course right still if you pre-intimate after that when the weekend batch starts you can jump to that weekend batch.

[00:53:57]It is going to start in November month.

[00:54:01]>> Okay. So you mean uh we can start from now and then go switch to weekend bench.

[00:54:06]>> Switch to weekend once you get enrolled to any of my batch right you can attend.

[00:54:10]For example, if there is any shift change, now you it's a 5 months course, right? For example, if any job shift change, right? So you want to attend the morning batch. So after that, you can pre-intimate, you can take that link for it and you can attend do that anyone, right? In this way, you can shift the change batch >> and also the weekend also, right? Once you get enrolled, you can attend that.

[00:54:35]>> Um, sir, I have one question. Um, hello.

[00:54:39]>> Yes. Yes. Go ahead.

[00:54:40]>> Yes sir.

[00:54:41]>> So you mentioned that you'll also be teaching MCP in agent uh in uh agent AI, right?

[00:54:47]>> Mhm.

[00:54:48]>> Or generative. So this MCP uh is basically operated in two modes uh standard input and um or socket or networking. So you'll be teaching both the modes or the standard input uh standard input output mode.

[00:55:02]>> Both the modes.

[00:55:04]>> Okay. Okay. Thank you sir.

[00:55:08]Okay. Um, any other queries you got? Anyone?

[00:55:13]Right.

[00:55:14]>> Just one query. Uh, sir, >> so instead of this MySQL videos, can we also have this Oracle video or like no instead of MySQL?

[00:55:31]>> Instead of MySQL videos, Oracle. Okay, that is the offer given by my Dasoft, right? It's from Dasoft videos only.

[00:55:38]Yeah, Dasoft Oracle.

[00:55:41]>> Uh, no, I don't think so. I I let I will let you know, right? Yes.

[00:55:48]>> Yeah, sure.

[00:55:49]>> I'll let you know.

[00:55:50]>> Yeah. Yeah. Yeah.

[00:55:52]>> Class will be like Monday to Friday. No, >> Monday to Yes. It will be from Monday to Friday.

[00:56:02]Yes.

[00:56:04]Okay. Any other queries before I wind up?

[00:56:09]Okay. Thank you all for your time. Thank you. Meet you tomorrow same time at 10:00 using the same link. Right. We discuss more from where we'll be discussing from where we are stopped.

[00:56:18]Yes. Good night.

[00:56:21]>> Thank you sir. Good night.

[00:56:51]Yes.

[00:56:54]Yeah. Good evening all. So before I start, can anyone please confirm whether my voice is clear, whether the screen is visible or not? Yes.

[00:57:04]>> Yes. Good to go.

[00:57:06]Fine thank you. So in yesterday we had some introductory session. I was just briefing you about data science and AI.

[00:57:12]So just I'll take five minutes what we discussed in the last session. Right?

[00:57:16]Data science like so everything deals with data is said right? Total world dealing with means flooded with the data user data. So data has got experience data has got future reference using this data and data science. we perform like different data analysis, uh data predictions, data classifications, data clustering, data recommendations, right?

[00:57:40]So yes, so we see blend of technique.

[00:57:43]The main technique what we see here in this data science is ML, machine learning, right? So I said data science combines math, statistics, programming, advanced analytics, artificial intelligence, AI and machine learning for extracting insights from the hidden in the organization's data in a shorter version. Right? I was saying data science is the art of turning the data into actions. Turning data into actions, right? Yes. One second.

[00:58:14]Data science focus on value, not the volume. Right? Yes, [snorts] data science I said it is rapid evolving at a rapid speed. So it is data science has been emerged from statistics and from data mining. It is all about the using the tools and using techniques and creativity to uncover the insights hidden within the data. So I said it is going to combine mathematics, statistics, computer science domain expertise to tackle the real world challenges in the varieties of fields.

[00:58:52]So data science it's means processing the raw data to solve the business problems and even making some kind of predictions about the future trends and the knowledge. Right? Fine. So generative AI. So generative AI, it creates a new content like text or images by learning a patterns from the data. Aentic AI, it focuses on making a decisions and executing multi-step actions to achieve the goal. In a simple way, right? In a simple way, I was saying generative AI is about the creation.

[00:59:30]Agentic AI is about action and decision making. So if you talk about AI, AI right the system that can perform task.

[00:59:41]AI means what? The system that can perform task that typically require human intelligence that require what? Human intelligence such as learning, problem solving and decision making. So this generative AI I can say it is a subset of AI. Generative AI it is a subset of AI. So designed to create a new original content.

[01:00:07]So it is going to read the patterns the structures from variety of data sets.

[01:00:13]Uses that knowledge to generate new outputs like text, images, code, music, all these things like chart, GPT, Delhi or some of those examples, right? And I say we have got a blend of techniques.

[01:00:28]So data science has got a blend of technique. The main technique what we see here is machine learning. Just I'll give some estate.

[01:00:37]Some of you are asking just I'll give some overview on this ML machine learning.

[01:00:46]So we perform predictions, classifications, clusterings, recommendations, patting, pattern matching. It's a blend of techniques. I said the main technique what we see here is machine learning.

[01:00:58]Yes.

[01:01:01]defining. Can anyone define machine learning? Yes, it is an approach. Yes, someone once again can anyone please confirm whether my voice is clear where the screen is visible.

[01:01:24]>> Yes sir.

[01:01:26]>> Yes. Thank you. It is an approach. It is an approach. I can say it is an approach where we provide where we provide some data patterns where we provide data patterns to an application to an application or a system. So I just doesn't want to enter into the technicals as of now. So in a simple way an approach where we provide data patterns to an application or a system and based on the based on the data patterns provided based on the data patterns provided the application the application or the system can predict can predict the things can predict the things in the automated fashion.

[01:02:30]Understand here it is an approach where we provide some data patterns to an application or a system based on the data patterns provided the application or the system can predict the things in the automated fashion. Right? Yes. For example, generally yes we provide historical patterns but generally we provide what historical patterns.

[01:03:00]See for example just I'll provide multiple example to make you understand.

[01:03:05]I'll take two variables. Taking two variables.

[01:03:10]So I'm taking X and Y.

[01:03:20]Taking two variables X and Y. See if X value if X value 1 Y is 10. If X is 2, Y is 20.

[01:03:30]If X is 3, Y is 30.

[01:03:34]So for four it is 40.

[01:03:37]per five it is 50. So in this way some x values and y values are provided. So simple examples this data is provided this data is provided to the application to the application right and it need it need to predict if x value is 27 what is y What is y value?

[01:04:18]Uh I want the sessions to be interactive. You can unmute and you can respond. Right? If x is 27, what is y value?

[01:04:29]1. If x value is 1, it is 10. If 220, 330, 440, 550, yes, >> 60 it will be 60. 7 it will be 70. Yes.

[01:04:38]270. Yes. Yeah. Most of you giving response. Right here what the model is performing here the model performs analysis it performs analysis and evaluates the model performs analysis and evaluates. So the relation between X and Y it relation relation between X and Y. Can you say the relation between X and Y? What is the relation between X and Y? So X Y is what? The relation between X and Y is what?

[01:05:27]H >> y = x into 10.

[01:05:31]>> Yes. Yes. Exactly. So y = 10 x. Can I say 10 * x? Y is 10 * x. Can I say for example if x value is 1? If x value is 1.

[01:05:52]If x value is 1, y = 10 x, 10 x means 10 into 1, 10 into 1 is 10. Similarly if 10 into 2 10 into 2 it is 20 10 into 3 30 if x value is 3 it is 30.

[01:06:21]So on so on so on. If you keep if x value is 29 sorry if x value is 27 is what? 270 right? 270. Hence y is 270.

[01:06:38]Y is 270. Understand the relationship between x and y. y is 10 * x. If you give any x value it will multiply with 10 and provides. Okay. The simple the simple thing. See check with other example. [clears throat] Example two.

[01:07:02]X is 1, Y is 10.

[01:07:08]If X is 2, Y is 14. Try to see this pattern. X = 3.

[01:07:16]What is Y?

[01:07:18]18.

[01:07:21]If x is 4, y is h. Anyone x = 5? Y is >> h everyone unmute and respond. I want the sessions to be interactive.

[01:07:38]>> 26 >> 26 if 6 30 if 7 3 4 8 but okay that's fine.

[01:07:48]Take some time and just try to say for this x = 500 then what is y?

[01:07:55]Then what is y?

[01:08:03]If x = 500 what is y? What is the relation? x relation x and y relation is what?

[01:08:12]>> Plus four every time.

[01:08:15]every time plus four.

[01:08:23]The model performs analysis and evaluates the relation.

[01:08:40]For example, if x value is 1.

[01:08:45]If x value is 1, y = can I say just someone said 4x, right?

[01:08:54]4x. Can I say?

[01:09:02]But you are getting only four, right?

[01:09:04]>> No.

[01:09:04]>> Four into one. How much I need to add?

[01:09:07]How much I need to add? See why you need to get 10? Understand carefully here 4 * x if you say y = 4 * x for example if you're saying you are getting four but how much I need to add to make it 10 here y = 10 actually I need to add six so 4 + 6 are you getting 10 but not for a particular input for all that input it has to if x value is two. If x value is two, what is y? Is it giving the correct four into 2 8 + 6 >> is giving 14 right? Even not only for one or two samples for all the samples it has to give that y = 4 into 3 + 6 how much? 18.

[01:10:08]X = 4 H Y = Yes Y = 4 into 4 16 + 6 what is that 22? So in the same way same way keep going for this >> we are humans we observing patterns but how system knows that this is the equation >> x = 500 >> x = 500 >> y = y = >> 4 into 500 >> 4 into 500 >> + left six >> 2006 >> what is that 2006 >> see if you're providing some pattern to that right it is going to evaluate but humans evaluation I was asking you evaluate what is the relation still so it has given that relation 4x + 6 so in the same way for example as you ask the query okay one more example here I'm taking two variables here also like two variables two variables for example there is some temperature sales temperature and sales temperature and sales right we say like direct proportional indirect proportional temperature is 27 31,000 the sales of the sales of something say product like cool drink the temperature and the sales of the pool 27 when temperature is 29 the sales increase to 33k 33,000 sales that is temperature 31 sales increased to 35k 33 became like 37 temperature increasing sales increasing 39K so in this way. Yes. Temperature 37.

[01:12:23]Assume it's summer. The temperature increasing the sales of the coolings also increasing.

[01:12:30]So this data provide to an application asking it to predict if temperature equal to 28. If temperature equal to 28 sales how much?

[01:12:53]So based on what it is going to predict based on the data that you have provided. If there is no data just if you say temperature 28 what is sales?

[01:13:03]So here what might be the sales? What might be the sales here? like to 32 or 33.

[01:13:09]>> 3K >> 32 or 3 something. It can be the sales because based upon what based on what basis it is providing this based on the data you are provided historical patterns the historical data the data which is already based on that it is going for example if temperature is if temperature is 24 what might be the sales will be like below 31 K means if temperature decrease sales also will be decreased temperature increasing sales is increasing right? Yes.

[01:13:45]So here based on historical based on historical data the model is going to be to get trained.

[01:14:04]So based on the historical data the model is going to get trained. For one more example just for example purpose.

[01:14:13]For example, Titanic ship.

[01:14:33]For example, assume total passengers total passengers they are traveling assume it is 1,000 in that total total dead 800 total survived.

[01:14:58]Total survivor 100.

[01:15:03]Unknown status.

[01:15:05]Unknown status.

[01:15:09]Unknown status is 100.

[01:15:14]One second.

[01:15:18]Insert this.

[01:15:28]Okay.

[01:15:30]See yes right Z. So using data science we perform analysis, predictions, classifications, clusterings, recommendations, pattern matching. So recommendations for the futures, strategizing for the future for all these things. So yes here unknown status is 100. Unknown status here.

[01:15:53]Okay, 800 dead 100 survived. But what about the 100 people predict what happened to this 100?

[01:16:07]Predict what happened to this 100.

[01:16:12]So understand carefully from the what you have already generally how will you evaluate this based upon the data what you got? Based on the data what you got for example you got 800 dead they will have some common features these people who are dead will have some common features will have some common features say just group them the features as X features X features features in the sense their name their age their traveling fair in their class. Which class? First class, second class, right? Okay. And people who have survived 100, they will have some common features.

[01:17:09]They will have some common features.

[01:17:16]I'll say that as Y features.

[01:17:22]the commonality I'm disc the people who have survived the 100 people in there in that I'm just extracting the common features saying this as y features using x and y features train the Model M model will predict Model will predict an unknown status. They'll predict on unknown status unknown 100 are there. Out of this 100, 90 has got X features. 90 has got X features and out of that N 10 has got Y features.

[01:18:32]Okay. What is the history is saying now?

[01:18:35]So the historic what is that data is saying? 90 has got X features, 10 has got Y features. Means this 90 people, what can you label them?

[01:18:44]What can you label them? Dead means the X features people have got X features they are dead. People who has got Y features they might be survived. So you're saying that 10 people might be survived from that 100 people unknown status. And depending upon the data you have provided based on the historical data.

[01:19:06]Okay. Fine.

[01:19:08]Similarly, if you see if you in the bank providing loans to the customers bank provided loans already provided loans base okay now some new customers are approaching for the loans to whom that loan has to be approved to whom that loans has to be rejected how it can Means for assume that thousand customers assume that thousand customers out of which out of the thousand customers right? Yes 600 people they're paying regularly they are paying the loans regularly.

[01:20:02]H assume that they have got some commonality. They have got some jobs with high salaries and whatever might be the other features we are saying that as X features X features. Similarly there the remaining 400 we say them as defaulters not paying not paying the loans. So just there were some Y features the people who are not paying this are commonality some features they doesn't have jobs even if they have jobs with less salaries. So assume new 100 new customers 100 new customers approach for a loan approach for a loan.

[01:20:55]What is the prediction out of this 100?

[01:21:03]Assume 30 has got X features.

[01:21:07]30 has got X features.

[01:21:10]Out of this 100, 70 has got Y features.

[01:21:19]What is your prediction?

[01:21:23]But this 30 has got X features means people who has got X features they're paying the loans regularly. So approve the loan for this. Approve the loan.

[01:21:34]And 70 has got Y features means reject the loan.

[01:21:45]Yes.

[01:21:47]Understood this example. So based on the data provided they are going to predict.

[01:21:52]For example, machine learning.

[01:22:00]It is an approach. It is an approach to train. It is an approach to train mathematical models.

[01:22:11]So before training observe before training model doesn't have any knowledge. Models doesn't have any knowledge after after training.

[01:22:35]After training model fit.

[01:22:42]After training it acquires that model fit.

[01:22:47]Model acquires.

[01:22:53]Models acquires the knowledge from the data.

[01:23:01]Here what are the model learns? Here what the model learns.

[01:23:13]It learns the patterns.

[01:23:16]It learn the patterns from what? From the data and performs what?

[01:23:24]And performs either prediction classification right here. What the model learns? It learns the patterns. Learning the patterns from data and performs prediction.

[01:24:05]Yes, sorry. Here what the model learns.

[01:24:08]It learns the patterns from the data and performs prediction or classification.

[01:24:17]Just once model fit once model fit is produced that has to be tested.

[01:24:32]That has to be tested.

[01:24:38]Once model fit deployed, Once model fit deployed into the application into the application right the application will become smart application.

[01:24:58]So once model fit is produced it has to be tested. Once model fit deployed into the application the application will become smart application. See uh what is the difference?

[01:25:12]Difference between normal application and smart application.

[01:25:22]Difference between normal application and smart application. Normal application.

[01:25:28]Normal application. Instruction based.

[01:25:33]Instruction based.

[01:25:35]Smart application intelligence based.

[01:25:49]So you train model requires model fit.

[01:25:59]Next, accuracy testing deployment.

[01:26:12]So here we train the we train a model here we train a model using data.

[01:26:24]The model acquires The model acquires knowledge called as model fit and perform testing or accuracy measurement.

[01:26:59]and later we deploy.

[01:27:08]So here we see okay so we have got different types of techniques available.

[01:27:14]So on which data which to be used we have got many here overfit under fit over fit under fit in a simple way for example a labor example overfit a birthday function a birthday party a birthday function you are expected 500 expected 500 So you arranged the food and everything for 500 but only 400 attended only 400 attended.

[01:27:54]This overfitit. Underfeit means for example expected 300 expected 300 but 500 attended.

[01:28:11]So okay machine learning is an approach as I was saying that it is an approach where we provide some patterns to an application based on that the application can predict the things in the automated fashion.

[01:28:23]So yes, one second.

[01:28:38]Just a second.

[01:28:49]Yesterday I was just discussing about the course overview, right? Yes. So initially I'll be starting with Python introduction and numpy and pondas for data analysis and data manipulations data visualizations using plotly and math plot lily and statistics was just discussing about statistics for data science machine learning techniques. Yes time series analysis statistics for data science. So this I was saying that statistics will help the data scientist understanding the data variability right and testing this hypothesis and summarizing this complex data sets and uh many many other things like dispersion measures sampling techniques, machine learning techniques, time series analysis, natural language processing, NLP, text mining, MLOps.

[01:29:44]So machine learning operations big data like pi spark and uh deep learning andn artificial neural networks CNN convolution neural networks right yes so in brief in depth we'll be discussing about these things tensorflow so it's a free opensource software library for machine learning and deep learning right boltsman machine and autoenccoder it is also a type of neural network that uses pro probabilistic approach approach to model data distributions and autoenccoder GAN generator to adversial more network RNN and GRU this all like deep learning techniques right this recurrent neural networks RN and recurrent neural networks GRU gated recurrent unit LSDM long short-term memory specialized type of recurrent neural networks in deep learning open CV right it is also open source Computer vision library plays important role in this deep learning also right and if we talk about generative AI from generative UI we see in brief in depth like I was talking about advanced AI with many topics on this AI generative AI right it's just I said creating new content like text images code and also prompt engineering advanced prompt techniques LLM APIs rack systems Vector databases and embeddings right LL LLM ops right log language model operations agentic system where you can see okay I'll come back to this but you can check this course content you can see this post content right yes one second hold on in brief in depth we'll be discussing so it is a full stack data science with artificial intelligence with advanced AI Okay. So total full total everything you'll be covering that full stack data science and artificial intelligence especially in artificial intelligence generative way and agent that two in brief in depth we'll be discussing. So you can just go through this data science introduction data science skills use cases data categorization types of data data formats Python for data science. So that introduction to this Python, NumPy for data analysis, pondas for data analysis, map plot lip for visualization, statistical methods for data science and everything machine learning time series analysis and forecasting MLOps and big data spark path with Python PIP right NLP and text mining extracting cleaning and pre-processing text deep learning tensorflow Convolution neural networks, Boltsman machine and autoenccoder, GAN, emotions and gender detection, introduction to RNN and GRU, LSTM, Open CV, generative AI, prompt engineering, advanced prompting techniques, coot, chain of thought, clear of thought and working with LLM APIs, building LLM APIs in longchain, developing rack systems.

[01:33:01]See from this 20 onwards first 2019 19 right for data science but from here onwards generative AI is going to start from module 20 to module 40 so that is I was saying in brief in depth we'll be discussing about these concepts building LLM maps developing rack systems vector databases and embeddings building M2 gen AI applications evaluating genai multimodel LMS MS LLM ops and evaluation from 31 to 40. You can see agentic agent architectures and design patterns working with long building with AI agents implementing agentic rack developing AI agents with free data multi- aent systems advanced development with autogen and building AI agents with no low code tools. See even this from 20 onwards to 40 we are discussing with this generative and agent. I was saying yesterday also I was saying just compare this post content with other training sites. So mostly if you see uh they'll be having one or one of page on this generative or aentic but we are discussing like 2240 model number 2240 in brief in depth that's what I was saying advanced AI. So if you come back even offerings from my site you'll be getting each and every session video you'll be getting each and every live session recordings even if you miss any session and go through the recording and just attend the next session for the better understanding clear soft copy of this notes notes will be clear very much you can see that each and everything you'll be understanding assignments and task to work and a WhatsApp group directory where you can directly have interaction with me directly where if you get any errors you can take snap of it and post in it interview questions around 200 questions right it'll be providing you yes dumar one [snorts] second live interactive hands-on sessions you can interact during the sessions you'll be kept in unmuted status at any point you can unmute and you can ask your queries and apart from these are the offerings from my set and apart from this and for this other things like the offerings from the dura soft set like the Python videos will be provided python videos like 100 plus videos on python will be provided myql videos will be provided right where you can just check with this anyone who does even I said I'll be discussing about python so around like two two weeks to 15 days I'll be discussing about this python pandas mplot lip numpy all these things uh but still if you want to go deep into that but that much is not required I'm saying the basic python knowledge is enough uh for these things and uh I'll make it clear about the timings of this patch before you go with the payment process first one week it will be at 10 p.m. 1 week to 10 days it will be at 10 p.m. it will be but later it will be at 9:00 p.m. Initially the class timings okay 9:00 p.m. to 10 p.m. 1 hour it will be but once we enter into this machine learning those things it will be for 1 and a half hour yes sometimes it also goes for 2 hours so let me make it the duration of the course duration is 5 months duration of this course right where you will be getting duration is five months so uh in that which includes two months for this data science this machine learning deep learning and around 3 months I'll be spending >> [clears throat] >> 3 months I'll be spending for generative agent. See understand carefully as compared with other training sites the other trainers other training sites in that 3 months they'll take 3 months time in that 3 months time again they'll go with Python they'll go with machine learn statistics they'll go with machine learning they'll go with the deep learning all this when you come to this advanced like generative identic they hardly spend one month time and just check the training sites that uh agent a generative but here explicitly for generative agent we'll be spending 3 months not for all this excluding all this right so we'll be getting full-fledged knowledge on this topics right so five months I'm saying clearly that's fine uh two months for data science topics and for this five 3 months on this generative agent where you will be getting the full advanced knowledge on this generative a and AI so five months so yeah all the prerequisites whatever are there everything I'll be make you understand and just go with the topic yes it's a it's a weekday batch right it's a weekend batch it's a weekday batch the classes will be from Monday to Friday sometimes I'll be taking on Saturdays sometimes I'll be taking on Saturday Raju right you can go with your query yes >> yeah I have a couple of queries uh one is uh how long uh you're planning to spend on Python.

[01:38:23]So, Python will be for 15 days. 15 days.

[01:38:27]Okay. And uh lang chain uh like uh one week.

[01:38:33]>> One week. Okay. Okay.

[01:38:36]And uh because uh I'm I don't have any knowledge in AI or ML or data science but I have I heard uh that uh statistics uh knowledge is required a bit at least.

[01:38:50]So we don't >> so everything everything the prerequisites everything I'm going to cover. So you can see >> so here even you can see module number six statistical methods for data science.

[01:39:05]>> Yeah. Yeah. Sorry I missed that wrong question.

[01:39:07]>> So initial initially I'll just go with this >> see uh Python for data science. Python module 2 is Python only. module 3 and four five three four five also related to python numpy pondas map first five six all these are the prerequisites right up to this statistics which I'll be make you understand then I'll go with machine learning but yes everything you'll be understanding clearly right step by step everything you'll be understanding 100% only thing is you need to attend the sessions regularly even if you miss any session you need to go through the recording and I want the session questions to be more interactive anytime you can ask your queries and even you have got WhatsApp group even the WhatsApp group creating from my side for all my batches I'll create a WhatsApp group where the students can have interact direct interaction from yes uh one last question I have so by any chance can I uh I mean can we meet you in person at the center or anywhere you would be visiting there >> uh I think later You have got uh you have got this WhatsApp number right me and you can direct interact with me.

[01:40:24]>> Okay.

[01:40:25]>> You can message me and you can I'll be providing the time right free time you can just interact.

[01:40:30]>> Ah sure got it. Thank you. That's it from my side. So people who doesn't have okay don't worry people who doesn't have knowledge in Python and I'll be just everything point to point I'll just discuss about this even in the tomorrow session just not not wasting the time just I'll be starting with the things from tomorrow so yesterday and today we have some introductory sessions I doesn't want to spend more time in this I'll just go with the things so people okay basic Python knowledge has So can I ask uh when are we starting with the actual syllabus?

[01:41:10]>> So from tomorrow itself right yesterday and today we have spent some time but with the basics. So starting with the Python right basics of Python.

[01:41:24]Okay. So tomorrow also people can attend the session same time at 10:00 using the same link. Tomorrow you can attend but from day after tomorrow the link is going to be changed. You are supposed to get enrolled for getting this videos notes adding to the WhatsApp group and also some free videos access provided by the dupa soft right.

[01:41:44]Yes.

[01:41:47]So everything uh everything the total data science and AI knowledge you'll be getting around with like three projects I'll be discussing during this course fine so same time tomorrow at 10:00 use the follow the same link tomorrow also but from day after tomorrow the link will be changed right any other queries before I wind up anyone uh actually this is uh apart from this session your new badge is going to start on 13th right 8:00 a.m.

[01:42:19]>> So for this morning some of the students they are asking for the morning that's for the morning batch >> so that will be at 8:00 a.m. only or any change in that time >> 8 a.m. only it will be at 8 a.m. only >> okay >> this I'm making it clear before the payments but yes it will be at 9:00 p.m.

[01:42:37]Fine. So, okay. Thank you all for your time. Thank you. Meet you tomorrow. Bye.

[01:42:41]Good night.

[01:42:58]Yes. [clears throat] Yes.

[01:43:09]Yeah. Good evening all. So before I start, can anyone please confirm whether my voice is clear, whether screen is visible or not? Yes.

[01:43:20]>> Yes. Good to go.

[01:43:22]>> Fine. Thank you. So just I'll take five minutes what we discussed previously, right? So it's full stack data science and artificial intelligence like advanced AI generative and agentic AI right so data science so as you combines math statistics programming advanced analytics artificial intelligence machine learning for extracting insights for that are hidden in organizations data so everything deals with the data I said data has got experience data has got future reference using data All these data science we perform data analysis, data predictions, data classifications, data clustering, recommendations as compared with the data engineering right data engineering right deals with the data management, data storage and data processing. So if I talk about data science shorter version, it is a art of turning data into actions, turning data into actions.

[01:44:23]So data science focus on value but not the value. So here data science it is about using how to use the tools the techniques the creativity to uncover this insights. So as I said what all the all the both things like math, statistics, programming the domain expertise right and computer science and uh processing the raw data into this raw data and solving the business problems and making some kind of classifications predictions and also making some predictions about the future trends.

[01:44:58]Yes, that's just a overview about this generate UI just creating a new content like a text or image by learning the patterns from the data. While if you talk about agentic AI like making some kind of decisions, making decisions and also some actions to achieve the goals, right? Yes.

[01:45:20]Yesterday I was just briefing you about ML machine learning where you provide some data patterns to an application or a system. Based on the data patterns provided the application can predict the things in the automated fashion. Two variables if you're providing this to this application right it has to predict if X value 1 2 as 20 3 as 40. So if X is 27 what is the prediction based on this data provided? 270.

[01:45:48]Similarly, some some kind of complex predictions. If X and Y values are provided in this way, X1 10 2 14 38 422 526 for 500 what will be the value? Some relation between X and Y will be evaluated that is 4x + 6 based on this it is going to compute.

[01:46:11]So if y is x is 500 it will be 2006.

[01:46:16]So based on the provided patterns even if it's take temperature in sales if the temperatures and sales values are provided tomorrow if you see what might be the temperature see for example not only this we got like classifications are available like for example last 4 days it is raining what is your prediction about tomorrow it may rain maximum from the last 5 days it is raining what is the prediction about means But the past pattern saying that okay maximum it may rain and what from last 5 days the temperature is 34 34 34 34 what will be the prediction about tomorrow around 34 but without providing any data patterns can you say what will be the temperature tomorrow you need to provide some patterns based on that it is going to predict here we got features labels we say right I was saying about one example yesterday like Titanic ship right now has got passengers like thousand passengers that how many are dead how many are survived and 100 unknown status based on this people who are dead people who are survived I'll take the features from there some common features out of 100 90 people have got X features means people who are dead they have got the features of people who are dead out of that 10 people have what why features the what features people of war survive.

[01:47:49]So from that past history is saying that this 10 people are servant 90 people are there. So these things I discussed yesterday and also like based on the previous patterns on the banks providing the loans right the previous history saying out of thousand customers 600 paying the loans regularly 400 not paying the loans regularly.

[01:48:14]Count of this people who are paying has got X features. People who are not paying have got Y features.

[01:48:23]So for the 100 new customers who are approaching for the loan, who has got X features, they'll pay the loans, approve the loan. People who have got Y features, the history is saying that reject the loan form.

[01:48:44]Yes, must be right that if any value is given for example some okay for example cinemas a simple example I'll give 5 years back 5 years back your age was 25 after 10 years predict your age after 10 years predict your Predicting your age after 10 years 35 means for example some age values are given the application can predict your ages okay for example for example as you asked one small labor example I'll give For example, a small kid. If you're taking a small kid, a small kid, his age is his age is 10. In which grade he is? In which class or grade he is?

[01:49:58]In which understand when he was in five years this age he was in first standard when he was in the age seven he was in the third standard.

[01:50:12]Predict the persons predict the person in which grade he will be when his age is 12.

[01:50:23]Eight standard right means if you are giving if this is the X value what will be the Y value? Eight.

[01:50:30]How it is predicting based on this? If these values are not provided can you predict okay tomorrow age for the person whose age is 14 what will be that class study. So the same this is a simple simple thing right in the same way we are providing some patterns here and we are making that predict the things.

[01:50:50]Okay.

[01:50:52]Temperature and sales values you are giving. Based on that it is going to predict that so machine learning is an approach to train some models. So before training there is they doesn't have any knowledge. After training they acquire knowledge and they can predict the things right. Yes.

[01:51:16]Okay. Once model fit means once the training once we train it it acquires model fit. Once model fit is produced we'll be testing it whether it is rating properly or not. Difference with normal application and smart application right normal application instruction based but smart application intelligence based. Okay fine we I discussed about overfitit and under fit. Overfeit means you have expected more but only less people have attended.

[01:51:49]You expected the result more but you got less. Underfeit means you expected less but you got the result more.

[01:51:58]Okay.

[01:52:01]Fine.

[01:52:04]So in the last two sessions we have discussed about this overview about this just today we'll be starting with the uh things.

[01:52:12]Okay.

[01:52:14]Fine.

[01:52:17]So here yes even lot of the lot of students were asking me about that language Python. So first we'll be starting yes what is that Nin right juggernaut I'm not going to discuss anything as of now about the technical so we'll be starting the thing step by step uh from ground level to that. So there are many life cycle is there step by step what is the life cycle I'll be discussing okay so generally why we go with okay we'll go with Python language as the implementation for this okay Python for data science generally Python for data science. You may ask why Python only used for the data science environment though other programming language like Java or other programming language not preferred. See Python is providing thousands of modules.

[01:53:23]Python providing thousands of modules.

[01:53:26]Python supporting 89,300 inbuilt modules for different environments. Whatever environment you are from whether data science or big data or database or networking or embedded systems or desktop any environment Python has got modules. The same way Python is providing many modules for data science environment like pandas, numpy, mattplot, lily, cps, key, kit, learn, tensor. This way many modules are there using that we can perform that operations and also it has got powerful libraries for different environments. Right? Yes, Python can say if you talk about this Python features, okay, I'll be coming back about it, right? So here, Python has to be installed.

[01:54:32]Python has to be installed in your systems.

[01:54:38]But I'll show you without installing also you can just work on this without installing Python also how we can work people who doesn't have knowledge on this Python right no need to install Python with installation without installation also generally two requirements two requirements to use two requirements to use Python in your machines Python in your machines See generally if you go in the company level right so you need to install Python apart from Python it doesn't work again one more thing you need to install only installing Python won't work you need to use ID use ids use ids various ids various ids integrated development ment environments various id supported by Python we have got many id in that jupyter notebook jupyter notebook is mostly preferred by is preferred spider ID is preferred vs code one more thing among all This Jupyter notebook, VS code, Visual Studio Code, right? These things compulsory. It is going to provide some flexible options here. This ID provides some flexible options to work with. Python code, various integrated development environment like Jupyter notebook, VS Code, Pycharm, Spider many are there.

[01:56:24]Still Eclipse is there, Pyave is there, Commodore is there. Right? Many other ideas. But for this data science environment, Jupyter notebook, VSport, Pycharm, Spider are mostly preferred for Python. But I said you may ask why I need to install all these things for working with Python as without installation also you can go. But as I was saying, right? Nowadays we are writing on cloud.

[01:56:59]We are writing on cloud. In cloud in cloud generally we have virtual machines. In cloud we have virtual machines and also like virtual storage.

[01:57:18]Virtual storage example of virtual storage.

[01:57:23]Google Drive example like the Google Drive.

[01:57:29]Similarly in the cloud similarly in the cloud we'll have machines.

[01:57:42]We'll have machines that two that two Python installed.

[01:57:48]That two with Python installed and the two with ID is installed.

[01:58:08]ID is installed in it.

[01:58:18]So that with ID is installed.

[01:58:24]We call it as Google collab. We call it as Google collab.

[01:58:32]So you can work with this collab. Google collab.

[01:58:39]So Google cloud. What is it? Google Cloud has Python installed. Google Cloud has Python installed called as Google Collab.

[01:59:01]Google Cloud has Python installed called Google Collab, right? Yes.

[01:59:15]Google cloud as Python install called as Google collab.

[01:59:20]Okay, people can work with Google collab without any Python installation. I'll show you with the collab and also how you can install in your local machine just hardly takes one minute for installing Python in your local machine.

[01:59:36]>> Okay. What Google What Google Collab offers?

[01:59:53]It offers virtual machine. It offers a virtual machine in which It has ID installed. It has ID installed in which Python is installed.

[02:00:29]What is that? In which water in which Jupiter ID? Jup which ID which it has Jupiter ID Jupyter notebook Jupiter notebook very [clears throat] simple like people who doesn't have knowledge on this Python or people who doesn't want to install this Python people might be working in your company systems where you doesn't want the installations to be done so you can go with this so if you have got just with a Google account you can just check with it. So just observe here, right?

[02:01:06]Um go to Google. Yes. Everyone you can try parally along with me.

[02:01:13]Yes. Say Google.com.

[02:01:15]Say if Google collab you say Google collab you say right. If it is asking so Google collab you will get collab.earch.google.com.

[02:01:26]Just say Google Collab. Click on it. If it is asking for this login and sign up login through your Gmail Gmail account, right? Yes.

[02:01:38]Okay. You'll get this open notebook can say open notebook. You get this option, right? If anyone so yes open notebook or it will ask for login or sign login right with your Gmail you can okay new notebook you can say new notebook you can say just notebook just like Jupyter notebook it is just it is loading a notebook you'll be getting here we can work with Python so you can see it here right you can see this is what see It is still not connected. See it is connect to a new runtime. It is saying connect to a new runtime. So if you if you execute any code cell if you execute any code cell it gets connected. It will be allocating some resources for you. For example, print good evening.

[02:02:38]Just print is a statement. Print is a statement which is used for out printing the output right or printing any statements. So just still it is not connected. Can you see your connect whenever if I want to execute this? If I want to execute this just click on this.

[02:02:55]Click on this run cell. Run cell. Now can you see it is connecting. It is connecting here. It is connecting and it will allow it will allocate some resources. Can you see disk and RAM? So, Python 3 version RAM and disk. So, it allocated some resources for you. It has allocated if you keep if you rest your mouse on this, it will show how much what is that RAM capacity and what is the disk capacity.

[02:03:23]It allocates some resources, right?

[02:03:26]You got the output also. Good evening.

[02:03:30]Yes.

[02:03:35]Yeah. So if you want if you want to open another cell if you want to write some other code it say here place code you can just click on this you want to work directly x as 10 y as 20 z = x + y no need to define any data types here right so just say print z that's it execute it you'll get this output 30. Output is 30.

[02:04:19][clears throat] Okay. In this way, you can just write your quotes and execute your quotes here.

[02:04:30]anything I want to just see how simple like even any graphs you want to generate any types of graphs you can generate here import for example import matt plot lib matt plot lib within this we have what a built-in functionality like piplot so python is providing many for visualizing for analysis for manipulations many models so map plot li alias name as p or plt I can say so just I'm importing for plotting a graph within that pip plot alias name or shortcut name I can make use of this plt dot plot for plotting a graph you require xcoordinate values 1 2 3 are the x coordinates.

[02:05:33]Ycoordinates values like 5 74 are the ycoordinates values.

[02:05:39]Some uh random values I have taken x coordinates and y coordinates. PL dot show can execute and you can run it. It will show that just observe here. Can you see a prompt got generated here itself.

[02:05:53]So just control minus control minus what now? See here for one it is five. Can you see for x coordinate one y is five and for two it is 7.

[02:06:09]For two it is 7. For three it is four.

[02:06:14]So in this way you can just generate a graph simple graph here itself.

[02:06:32]Instead of PL if I say B bar graph will be generated.

[02:06:37]Instead of PL if I say hi S histogram will be generated. Instead of pl if I say P I p will be generated. Instead of pl I say scatter scatter plot stack plot any type of graphs you can generate. It's very simple here the Python statements not to worry people who doesn't have knowledge on Python not to worry it looks like normal English statements without any knowledge on Python I can make you understand this Python statements because they look like normal English statements just observe one more example same this just you can control plus control plus I'm increasing the font import Matt plot for plotting within that pip plot we have got a built-in functionality as P alias name instead of using this very big name in my statements alias name as P or P plt I want to generate I want to generate a piraph so pigraph I want to perform some kind of analysis I want to perform some kind of analysis so for example just a sample example I'm showing you to make you understand without any knowledge on this Python also you can understand the statements I'm saying just sample example course I'm showing you how to generate graph how simple generating graphs the importance of models in Python why we use only Python for implementing this data science this machine learning deep learning all these things right and for example in a simple way I want to analyze is how I'm to just analyze how are the sales of a particular product particular product like mobiles I'm taking some names I'm taking Samsung okay Samsung vo some names of the mobiles in this How are the sales? For example, last month's sales, August sales, Samsung 45 got sold. Vivo 10 got sold.

[02:09:04]Oppo 15 got sold.

[02:09:08]Yes, OnePlus 30 got sale. These are the sales of just I want a pigraph. Please generate a pyraph. I'm saying a pl. Pi I said for plotting a normal graph plot function for generating piraph P I E A pi graph will be generated. Now pigraph based on August sales. based on August sales.

[02:09:33]I want a pyraph based on August sales.

[02:09:36]And I want some output. I want some labels in the output. Labels. Labels.

[02:09:41]Nothing but the mobile names only. You take it as the labels.

[02:09:46]Okay. Labels as the mobile names. PLT.

[02:09:49]Do show. That's it.

[02:09:53]Execute it. A pigraph will be generated.

[02:09:57]Just a pigraph will be generated. Can you see with the different colors?

[02:10:02]Can you see with the different colors a pigraph generated?

[02:10:12]Yes.

[02:10:14]Can you see this? So more sales are for Samsung. Can you see the blue colored?

[02:10:19]Samsung has got more sales. OnePlus has got more and Oppo and Vivo. more for Samsung. Next, OnePlus, next Oppo, next vivo. So just check it here. Here itself. Can you see the statements?

[02:10:33]Easily understandable. Mobile names. The sales of August. A pyraph based on August sales labels equal to mobiles.

[02:10:44]Just I'm showing you some sample example where to generate a graph using Python.

[02:10:51]And the two is the statements.

[02:10:54]Is it understandable or not? Do you see any complex word or statements here?

[02:10:58]Nothing. Right.

[02:11:02]Okay.

[02:11:06]Any types of graphs you can generate in Python? Very simple, right? Yes.

[02:11:12]Here one more graph with two lines just we generated a graph. So anything here in this collab we are using collab right where no need of any python installation no need of any ID installation generally we use Jupyter notebook mostly for this data analysis.

[02:11:30]Yes, even you can make use of any Linux related commands also. Here we say like ls if you say and execute ls command list s and if you say pwd all this will work pwd present working directory execute it where currently s / content in this way it is going to any any command you can execute here if you want to upload anything you want to upload anything here can you see files can you see files here if you click this file You'll see that upload option. Upload.

[02:12:09]Upload to. Yes.

[02:12:12]Just if you want to click anything. If you click anything, any file.

[02:12:18]Any file. For example, any file. I'm loading.

[02:12:24]One CSV file. I'm loading.

[02:12:27]Loading. Right. Yes. A file. You can see a file got loaded here. It's a pmp1.

[02:12:34]CSP. So here like a even if you want to work with anything all this working with pandas import pandas in pandas we create a data frames. So just you can check with this how I'm creating a data frame within a single line just without see I want to read this CSV file and I want in a structured format in the form of rows and columns in a tabular structure I want to create a data frame so just df data frame using this pandas we have got a functionality read csv and what is Just now I seen pwdt means whatever you are loading everything will be stored under this content /c content. So /ashc content that is your current present working directory within that emp1 csv or even if you're not giving also by default it will take it emp1.csv CSV control plus increase the font just I'm showing you the statements and how they look pandas dot otherwise even it is the shortcut name or you can say import pandas pandas is already available pandas dot read cvv file emp1 csv print df can say data frame it's going to be created And it is going to display the data of that file. Can you see the sample employee e ID e name salary department number city just can you observe the statements one just I imported pandas and I'm saying within the pandas to read a CSV file we got to read csv reading a CSV file the file name which you want to read this is the functionality print df data frame got created in this way pas data frame will be created very simple for uploading and just creating data frames So everything looks simple here right python coding or if you want to work with pandas numpy m plot li for example just I'll upload one more file just for your practice one second just observe here upload option is there here upload if you want to upload an excel file For example, I got an Excel file, empxlx, Excel file, uploaded an Excel file.

[02:15:26]Very simple. The same way anything uh it's very simple to code. I'm just copying the same code.

[02:15:34]Copy the selection. Paste it here. Ctrl + V. Import pandas. To read a CSV file, read CSV is there. To read an Excel file, understand Eel. Just I was showing you some sample codings. How simple they look. Read Excel. To read an Excel file, uh what is the Excel file name? Uh EMP.

[02:15:57]XLSX.

[02:15:59]That's it. Here also data frame gets created. Simple, right? Sorry.

[02:16:08]EMP EMP right not empir.xlsx emp.xlsx now now you can see a data frame got created from this a table structure in this way panda's data frames you can create you can perform different operations once data frame is created we have got many data frame related operations from this again I want to you may ask okay from this data frame I want to create okay from this data frame I want to create a CSV file an Excel file a JSON file dataf frame dot to okay to CSV I want to create a CSV file uh for the example employee 1 CSV uh one data one CSV file I want to create in double quote CQ in double code CQ. Now data frame dot to Excel to Excel data frame convert to Excel and create an Excel file employee to employee to do XLSX.

[02:17:32]Next data frame dot I want to create a two JSON to JSON right where is it to JSON I think I can also create a JSON file employee employee 3 dot JSON I'm creating once see I created a data frame from a particular file later from a data frame I can create a CSV file Excel file JSON file Just execute and check. Just execute and check. Just at the left side, can you see now files will be generated here.

[02:18:10]Uh now can you see employee 1 dot CSV, employee 2.xl, employee 3.json.

[02:18:18]How simple. If you want to see this Excel file, you can you can download and you can see it. Okay, these are the files. Simple generating files. Excel files or from an Excel file creating a data file. Simple just I want to show you how simple the codings work here and also generating graphs using map plot lib model using pandas how to create a data frames like and also you can work with Linux commands. I said in this collab you can work all these things without any installations.

[02:19:01]Okay, for example, I want to display the first two employee records data frame of for example DF of zero 1 second.

[02:19:31]Okay. Uh DF of I want only DF of E name only employee names only. I want to print from this data frame. Only employee names employee names only it is printed.

[02:19:55]So in this way the total everything will be simple just I was showing you without installing any Python without installing any Jupyter notebook you'll be getting the same Jupyter notebook kindru ID structure here in collab no need of any installation can work with it otherwise if you want to install also it hardly takes 1 minute for installing Python in your local machine it hardly takes 1 minute for installing Python in your local machine you can install it and also ID installation also ID installation also right it hardly takes 5 minutes for installing it's not that much complex but so yes uh uh one second Okay. So just I was briefing you about this just I doesn't want to enter into more technicals and from last two days I was just briefing you about introduction to this machine learning data data science AI these things. So if you see in brief in depth from the ground level to the advanced level. So here it's a five months course. I said it's a five months course where we'll be discussing about total full stack data science the prerequisites everything will be covered 2 months for this data science 3 months for this AI 3 months with this AI in that generative and agentic AI see once once again I am specifying 3 months explicitly for this not like other training sites not like the other trainers in the artificial intelligence again uh this uh generally in other training sites other they'll be discussing again Python statistics machine learning deep learning all this uh text mining all this will be discussed not like explicitly I'm spending time 3 months for AI means generative agentic AI advanced AI apart from this two months for this data science so data science starting with data science introduction why data scientists are in demand So data acquisition types of data data data formats quality quant data quantity and here so as I said Python for data science Python basics uh like collections functions dictionaries file handling right collab and numpy python for data analysis pandas python for data analysis numpy pandas for data analysis and data manipulations Matt plot lip for data visualizations statistical methods for data science. So you need to have some statistical knowledge that's why I'm giving the prerequisites of the statistics also descriptive statistics inferial statistics dispersion methods types of sampling correlation analysis machine learning. Next module 7 is machine learning and all the machine learning techniques and with implementation time series analysis and forecasting MLOps.

[02:23:13]MLOps right? Yes. MLOps introduction and big data like Pispark. Big data like Pispark NLP and text mining. Just a second. Just a second.

[02:23:53]Yes, big data pispark.

[02:23:55]So yes, pispark is a very high speeded processing technology for processing huge data at big time. NLP and text mining, right? Yes.

[02:24:08]Extracting, cleaning and pre-processing text. So it allows Okay. NL here we see like NLTK natural language toolkit that is a suit of libraries and programs for symbolic and statistical NLP right and also like deep learning where I'll be discussing about ANN artificial neural network convolution neural networks tensorflow it's a free open-source software library for machine learning and deep learning right it's developed by Google right and Convolution neural networks, Boltsman machine, type of neural network, generative GAN, emotional and gender detections, RNN and GRU, right? LSTM, yes. Long, short-term memory, yes.

[02:25:04]Open CV and from 20 onwards module 20 to module 40 generative A and aentic AI my discussions are going to be on generative A and aentic AI okay in brief we'll be discussing about generative AI the prompt engineering techniques advanced prompting techniques working with LLM APIs developing rack systems vector databases and embeddings building end to end gen AI applications eval evaluating gen applications in enterprise market multimodel LMS LM ops and evaluations and from 31 to 40 agentic I'll be discussing agentic architecture and design patterns working with longchain building AI agents with longchain topics implementing implementing agentic rack developing AI agents with feed data multi- aents systems with long graph through AI advanced agent development with autogen and building AI agents with no low code tools. Yes.

[02:26:16]So today just I was discussing about this Python for the implementation of this and just I have a showcase of what will be discussed. Yes. Tomorrow onwards we'll be start with the actual things.

[02:26:28]Yes. And also I was saying right what are the offerings from my side.

[02:26:35]So tomorrow the link will be changing you are supposed to get enrolled for getting the link. So you can see in the chart one second can you see in the chart if you if anyone if you have missed the sessions of demo one and demo two can see that the links are given for it. Not a problem. From tomorrow onwards daily the videos you'll be getting to your Google, right?

[02:27:00]And notes to your mails and uh tomorrow a WhatsApp group will be created. You'll be added to the WhatsApp groups.

[02:27:07]Yes. And the link will be forwarded to your mail tomorrow. It's a paid link.

[02:27:11]You are supposed to get enrolled and you can see uh the contact details of this online if you got any concerns related to this. And uh yes and this payment details payment details are provided here. Can you see that? And don't forget to share the screenshot of your payment to that mail id there forward. Then only they'll share the link with you. Don't forget to share the screenshot of your payment to that mail id that is given. Then only they'll forward the link. Most of the students they get enrolled they won't forward that screenshot.

[02:27:51]Yes. and the Google pay and phone pay details are provided and the offerings from my side as I'm saying okay this all that right yes session videos recordings you'll be getting each and every session video you'll be getting clear soft copy of these notes you'll be getting everything in a assignment sentence to work and a group for technical discussions where you have direct interaction with me interview questions around 200 interview questions live interactive hands-on sessions and even the things if you want to work parallelly you can work along with me during the sessions you can have hands-on experience you want to do parallelly and apart from this this is the offerings from my side and from duro software they are going to provide around 5,000 rupees access to the videos of python and myql around 100 plus videos will be provided You can go through that if you want more command on this Python and MySQL videos around 35 videos for this database who doesn't have knowledge on this. And let me make it clear about the timings of this batch. For the first one, it will be at 10 p.m. Later it will be starting at 9:00 p.m.

[02:29:11]So initially it will be for 1 hour 9 to 10. But later when we start with machine learning those things it will be for 1 and a half hour. duration is five months, two months for data science and three months for this AI in brief in depth will be discussed right see once again I'm saying it's an offer batch just for 20,000 because previously data science was taken separately was taken separately so data science will be for 20k AI will be for 20k it's totally 40k if you're going one of separately but it's offer b just given for 20k where we'll be discussing each and everything from the ground level to the advanced level. The prerequisite is also covered and also pi spark like big data is covered. Python is this modules like numpy pondas are covered right and uh generative agent in brief it is discussed right. So any queries anything you can ask during the session itself you can unmute you can ask then and there and even maintaining a WhatsApp group where you can directly have interaction or you can take snaps of the errors if you got any fine any queries you got anywhere but any queries you got so so just get enrolled So tomorrow there is a change in the link. So to get that link, get enrolled and to get this video access free video access and also to get the videos notes and to get added to the WhatsApp group right just to get enrolled for tomorrow's session right so that is what the offerings from my side and what's the course content what I'm going to discuss right and around the three projects will be discussed I said in this this.

[02:31:10]Yeah. Today is the Yes, today's the last.

[02:31:18]Yes.

[02:31:21]Any other query start anyone?

[02:31:25]So, okay, we'll start from tomorrow.

[02:31:27]From tomorrow, we'll be starting with natural things. Thank you all for your time. Thank you. Meet you tomorrow. Bye.

[02:31:33]Good night.