Data Science FREE Master Class | 10:45 AM

Hinzugefügt: 2026-05-06

[00:00:03]Good morning.

[00:00:16]Centurion GC.

[00:00:34]How we take third Good 10 minutes.

[00:01:34]off.

[00:01:37]So basically basic data science, data analytics, AIC First of all, online videos, data science, data analytics question guidance.

[00:02:40]What a normal story.

[00:02:46]It's a kind of a So basically two recently access Google Drive.

[00:03:45]particular platform desktop view.

[00:04:13]Ed Google same structure that is a very collaborative collaborative tool for enterprise level development GitLab.

[00:04:48]Okay, come to the point.

[00:04:53]Then simple command.

[00:05:04]Simply resolve this.

[00:05:28]Linux language different.

[00:05:48]Second request how is impacted.

[00:06:21]is seriously kind of AI can't replace us.

[00:07:02]Optimize complete development tools development different tools.

[00:07:32]testing different management machine learning architecture means management team.

[00:08:31]literally suggest the I don't know about all and also datiding you the information how works web scrapping What concept webtm structure differentological department, Indianological Department basically departmentalis.

[00:10:57]humidity.

[00:11:00]Same scenario.

[00:11:17]Mumbai past let's Let's say how you are basically analyzing that analyzing through data and the same scenario like most topatches machine learning model simply LB of height.

[00:13:27]What is our importance as a data scientist?

[00:13:46]But we are trying to be Obviously, that can be done through AI, but everything cannot be done through AI.

[00:14:35]everything in the sense of major complex mathematical calculations, major complex algorithms.

[00:14:42]Huh.

[00:14:47]But what is the line between First of all, in next 10 to 15 years, AI cannot replace everyone in the engineering field.

[00:15:17]In engineering field, let's say but do you know how to implement that code in a complete enterprise level enterprise level Each small performance can create a complete enterprise level applications and we are built for enterprise level.

[00:16:21]Simply learning to model training testing deploy without it will take you 3 to four months but obviously advanc level.

[00:17:42]As a fresh The same structure performance.

[00:18:22]Basic data set.

[00:18:43]Excel sheet.

[00:18:48]Basic information accident.

[00:19:10]State person injured road type accident.

[00:19:37]Better for dash.

[00:20:04]Total 3,83 records 36 then killed injured Accident.

[00:20:40]So basically representing data from table way to a visual way is our work and more efficient way. Normally vehicle Not report top accident.

[00:22:16]I'm engineers.

[00:22:18]User department hotspot analysis let's say health department service commiss How everything is linked, National Highway Authority of India.

[00:23:28]already department already. These kind of applications are running in different states.

[00:23:46]How how useful it will be and relate table.

[00:23:57]How it is user friendly and how informative it would be the top states by accident.

[00:24:17]Top states complete data center of research.

[00:24:51]Okay.

[00:24:58]information and most important thing.

[00:25:04]These are basic understanding.

[00:25:19]These recommendation sections are basically from machine learning models.

[00:25:30]machine learning model 54 accidentally toprimp improve signal.

[00:26:02]Then local traffic control traffic control and measurably awareness boards.

[00:26:15]These improvements recommendation can be personalized also.

[00:26:21]Improvement recommendation at the same install speed breakers and stronger speed monitoring. These recommendations are bearing as per the cause, release score and majorly type of accident.

[00:26:59]This is called real time dashboards.

[00:27:04]As per real time data, it will behave as per the current information and reflective the output. Basically, district 23.

[00:27:42]Top 23 risk distribution highly for accidents.

[00:28:05]Then hotspot finds that we can do with a map.

[00:28:24]information.

[00:28:27]Let's say how it becoming real time. It is becoming real time only due to these kind of dashboards, these kind of graphs and charts.

[00:29:10]particular information.

[00:29:26]These are basically We can improvise it like GPS implementation.

[00:30:04]machine learning.

[00:30:06]How machine learning predictions basically working?

[00:30:22]Total target accident model.

[00:30:44]Now particular location hotspot warning as datetch.

[00:31:15]Don't worry about it.

[00:31:20]After predicting it will give me outic.

[00:31:53]What a complete application.

[00:32:22]End result.

[00:32:35]means you can develop your own project.

[00:32:39]You can explain your own project and the most important thing you can explain your logic logics.

[00:32:52]for YouTube.

[00:33:19]Data science analysis, machine learning structure database. SQL format. Yeah. Any kind of storage unit.

[00:33:56]First of all, data science is a combination of data, data understanding, data analysis and machine learning.

[00:34:09]basically machine learning different problem.

[00:34:37]Let's say information database relational database non relational database value.

[00:35:20]data extraction relational database server. for opt.

[00:36:13]Select DML basically database let's say chapter master different database information.

[00:36:52]Let's say example different city collection.

[00:37:38]Environment analysis department.

[00:38:00]Second step.

[00:38:05]Second step.

[00:38:20]Basically collection process ID.

[00:38:51]data science and cloud cloud basically and everything will be stored in cloud.

[00:39:20]machine learning. No problemat Google different code also.

[00:39:42]We can work in Anaconda environment also.

[00:39:51]Note spider Google notebook.

[00:40:24]everything but difference cloud environment local environment local environment local operating systems and RAM so integrated completely but cloud environment first. Second thing, everything is pre-installed in Google.

[00:41:05]You need to install everything.

[00:41:10]And third, most important system.

[00:41:22]There is no back.

[00:41:40]Google is restricted within the boundary itself.

[00:41:53]local system, but always we need to back it up.

[00:42:13]Simple free.

[00:42:36]All those LLMs you can use in your laptop for your performance improvement.

[00:42:53]Simply say particular data.

[00:43:18]Give me a complete information of this data set.

[00:43:22]information then structure.

[00:43:41]This is not data security wise means which is secure. Most important thing why we cannot upload our data to the public platforms cloud these are public platforms.

[00:44:26]Most important reason in enterprise data need to be private.

[00:44:48]as a data scientist.

[00:44:52]We work on data.

[00:45:07]Complete information overl kind of next.

[00:45:25]Next hotspot max.

[00:45:49]Python arbiton division.

[00:46:23]simplest language.

[00:46:25]>> Python slow Java is actually Java.

[00:46:36]Python Python slow.

[00:46:42]But why Python?

[00:46:52]syntax.

[00:46:56]Okay.

[00:46:58]See banking systems complex logic be complex Simpler syntax.

[00:47:41]Firstchech.

[00:48:08]How basically Python because machine learning data science analytics related all libraries are pre-built in Python basically function.

[00:48:29]What is a function?

[00:48:34]Definit I'm explain multiple times.

[00:49:21]User function function.

[00:49:49]A B value.

[00:50:00]Return a B.

[00:50:12]variable concepts.

[00:50:30]Addition machine learning.

[00:50:45]said this is something forest.

[00:51:12]Body reusable block already.

[00:51:58]Let's say library.

[00:52:15]Import library.

[00:52:30]Eight type of block of structure already data. Numeric data. Tabular data, numeric.

[00:53:22]PG basically data manipulation library means data analysis concept operation row operation function. Everything is a function.

[00:54:18]document user guide reference developer guideated versions Then basically information Pandas are nonpelment.

[00:55:49]data read process data understanding process data frame series operations column operations locator I locator sorting concept basically understand these information.

[00:56:38]Then analysis.

[00:56:49]Let's say Number eight concept missing.

[00:57:37]information related to your personic information.

[00:57:52]Let's say 8 A missing information MC M& with group edge. Male edge class Let's say but same age let's say 21 age male weight average 60 65 average Max to max 50 index.

[00:59:57]Obviously, groupmale female male missing clean years.

[01:00:58]Wrong information.

[01:01:20]Next floor analysis exploratory analysis data analysis.

[01:01:30]This dashboard exploratory data analysis parting always open for any kind of comments.

[01:02:12]Okay.

[01:02:18]Data visual conceptual understanding feature extraction feature engineering.

[01:02:49]These are some concepts.

[01:03:01]CC glass.

[01:03:14]Okay.

[01:03:20]Pythonic set Everything you can develop.

[01:03:59]election.

[01:04:19]Okay.

[01:04:31]Exit population.

[01:05:01]analysis statistics.

[01:05:20]Congress.

[01:05:29]Let's say count to visualize some highest accidental pra let's say okay graphic Dashboard create different graphs.

[01:07:10]Everything search plot types different plots different different plot information such as histogram, box plot, piecular Dashboard dashboard accident.

[01:08:46]accident.

[01:08:48]Okay.

[01:08:57]Over speeditting of the accident.

[01:09:13]Top 10 that is the need of data visualization.

[01:09:30]Then functions then basically univariate analysis analysis multivariate analysis of time series analysis reports by day report means information Tuesday analysis.

[01:10:30]coins.

[01:10:38]Good.

[01:10:53]Pythonic library for web developer.

[01:11:09]analysis and analytics with machine learning.

[01:11:32]You have your own project in worldide.

[01:11:51]Obviously you can simply say your URL everything you will get your job rather than Google there is huge difference and most important thing machine learning and generative generative API means data command Applica predict dictionary. concept.

[01:13:51]JavaScript object How to create? How to link API?

[01:14:37]machine learning project then we will wind up machine learning concept What is machine learning?

[01:15:20]>> Using past data.

[01:15:32]Okay.

[01:15:45]like that.

[01:16:06]Learning through particular objective and that level.

[01:16:13]forced.

[01:16:40]Machine learning machine learning.

[01:16:45]Machine learning is a concept using some data and say that is machine learning. Simple.

[01:17:12]In supervised learning mostly we provide data and label both input and output It's simply actionoriented.

[01:18:00]the state basically reinforcement learning in machine learning that is reinforcement learning processing Next time that events learn, unsupervised learning.

[01:18:56]Clustering concept learning.

[01:19:19]learning action oriented oriented reinforcement learning backtracking back propagation to start deep learning. Deep learning language processing networks start then we can be also machine learning model.

[01:20:08]First it is divided into two type of problems regression and classification.

[01:20:19]Uh let's say dollar price as per data price analysis report basically data Usually increasing dollar price.

[01:21:20]There is a continuous process.

[01:21:34]If it is continuous and numeric under These are classification models. We need to classify among different class.

[01:22:13]The classification binary classification Multi classifications regress max analysis classification Multiclass classifications that we are predicting a numeric value model classification logistic regression Tree based model.

[01:24:08]Let's say let's say 40% let's say everyone Proper information 40%.

[01:24:56]Let's say let's say weather Yes 100%.

[01:25:17]No 0% or vice versa. This 0% this is 100%.

[01:25:30]We take decision through some conditions.

[01:25:37]And that decision taking goes on until we gather information 100% or either complete true or complete false.

[01:26:18]post pruning concept Maximum no pre-unch.

[01:27:13]posting model forest.

[01:27:25]Random forest model.

[01:27:56]Data science is basically a concept of collecting data, exploring data and gathering information from data.

[01:28:06]Say information through analyzing data and predicting outcome of that data.

[01:28:12]Clear? data collection, data prep-processing, data cleaning, exploratory data analysis, dashboards development, model split, data splitting, model training, model testing, model evaluation, performance improvement, process prediction, machine learning related variance of important important concept related to statistic and related to data visualization and analysis data understanding, data generations, data collection and other models rather than creating our own model pre-built model already pre-built ji models or models and all integrated structure data collection.

[01:29:59]Clear as a data scientist.

[01:30:09]First of all, you can join any industry as a data analyst, visualization engineer, engineer.

[01:30:28]Everything.

[01:30:47]This is the power dashboard without Max found data collection data manipulations.

[01:31:35]SQL and web scrapping data collection then data understanding and data visualizations using pandas numpy and macro lily data visualization in powerbi creating dashboards using stream and pandas nump then after this we will enter to machine learning machine learning different phases need of machine learning machine learning algorithms practical data Then machine learning complete machine learning. Each machine learning model Then prediction complete machine learning. Then it will be a complete data science and analytics project.

[01:32:58]How to create API and data collection?

[01:33:05]Clear.

[01:33:27]Everything by complete project and clear data analysis analytics project then clear and everything Clear and most important thing each month two times data scientists belong to Microsoft they're actually working there as a data scientist or analyst anything hours mostly Saturday monthly two times connect you can interact with him directly or her directly about any kind of queries related to career related to development.

[01:35:04]Okay. Okay. Now the basically initial information.

[01:35:30]That's why information Simply conclusion and everything.

[01:36:03]Next she is handling all the preparations. Basically, data structure questions previous Okay, thank you so much.

[01:37:15]Okay.

[01:37:20]Gest