Applied Data Science With Python Full Course 2026 | Applied Data Science With Python | Simplilearn

Every time an app predicts customer behavior, a company studies sales trends, a bank detects unusual transactions, or a business decides what to do next using data, data science is working behind the scenes. And one of the easiest and most powerful ways to start learning data science is with Python. Hey everyone, welcome to this applied data science with Python course by simply. In this course, we will start by understanding how Python is used in data science and why it has become one of the most popular languages for data analysis, visualization, and machine learning. You'll be learning how to work with important Python libraries like numpy, pandas, mattro, lip, cboard, scikitlearn, which are widely used by data professionals. We'll then move into handling real world data where you'll understand how to create arrays, work with data frame, clean data, analyze pattern, manage missing values, and prepare data for better insights. After that, we will explore data visualization where you will learn how to turn raw numbers into clear charts and graph that make data easier to understand. We'll also cover important foundations like statistics, linear algebra, categorical data handling, and basic machine learning concepts. So you can understand how data is used to make predictions and support business decisions. And by the end of this course, you will have a strong practical understanding of how Python is used in applied data science from data handling and analysis to visualization and machine learning basics. So without any further ado, let's get started. Also, if you're ready to take your skills in Python and data science to the next level, check out this SimplyLearns data science with Python course. Now, this course is perfect for anyone who wants to master Python programming for data analysis, visualization, and machine learning. You'll be learning how to work with key Python libraries like NumPy, Pandas, Mattplot Lib for data wrangling and analysis. You'll dive deep into data visualization, feature engineering and statistical models that are so important in the world of data science. Plus, you'll be getting hand-on- experience through real world projects like sales analysis, marketing campaign analysis, and many more. And upon completing the course, you'll be earning a course completion certificate from Simply Learn, which can boost your career and showcase a new skills to the employers. Check out the description below for the link and start your data science journey today with Simply Learn. Before we get started, here's a quick beginner friendly quiz for you. Which Python libraries mainly used for data manipulation and working with data frames? Your options are mapplot lib, pandas, seaborn or scikitlearn. Let me know your answers in the comment section below. Introduction to data science. How many of you have heard the term data science and what do you think data science is? This particular lesson you'll be able to explain the basics of data science and its applications to derive meaningful insights as responded by Wuni and uh Siki Pan list the steps of data science process to solve problems systematically and explore the Python packages for data science to efficiently perform data analysis, data manipulation and data visualization. Also, you'll be able to describe the types of plots available for visualization to communicate data insights and trends effectively. Let's get into the introduction. Yeah, what is a data science? You need to understand data science is a multi-disiplinary field. that uses scientific methods, processes, algorithms and systems to derive meaningful insights from structured and unstructured data. Right? So you need to understand what is a structured data and unstructured data. structured data. You can see as I found many participants has gone through the SQL course. What are the data stored into the SQL tables in the form of rows and columns? Right. Very good. Very good. What are the data stored into the database tables in the form rows rows and columns like it may be a employee details it may be a e-commerce order details it may be a bank records student details right which contains uh column names like for if if it comes to bank records name account number balance address email id mobile number as a column names and all the records you'll be having right which is called as a structured data right what is the unstructured data unstructured data which doesn't have any particular structure the data is in the form of a text images or emails which usually don't fit neatly into the tables Like you can take the example of social media posts which would be Insta comments, tweets, images, videos being shared on the social media. These all are called as unstructured data. Using a search engine or making a purchase on Amazon provides valuable data to data science-driven software systems operating in the background. Data on interactions with online platforms is gathered to understand user preferences and suggest search results or items to buy. When you make a purchase, for example, if you are trying to buy a laptop, immediately you'll be getting some kind of suggestions, right? What kind of suggestions? People who has bought the laptop also bought a wireless mouse, the earpods and backpack bag and so on. How you get these differences? Earlier people who has purchased the laptop might have purchased those products and it is analyzing the product you have you purchasing and relating the same product buyers data and finally suggesting you some results. That's how the data on interactions with online platforms is gathered. So by this what you came to know structured is organized easy to search and uh stored in database whereas unstructured data is free form scattered and harder to analyze but contains valuable information. Here you can see I was telling you data science is a combination of methods, scientific methods, process, algorithms and systems. Right? To derive meaningful insights there is a raw data. From the raw data I want to bring out for example I would like to give you an example related to purchases on e-commerce due to the some sale happening on the e-commerce website millions of people are purchasing the products soon after the sale completed I'll have good amount of data in hand and I want to analyze is the purchase trends like how many users has purchased mobiles. Okay, I'll get some amount of data right and over the data I would like to perform further analysis age group analysis. What is the age group of people purchased different mobiles? And you might be getting the people with age group of 20 to 25 might have purchased uh some XY Z brand and uh people from 26 to 30 might have purchased some XY Z brand, ABC brand So that's how you keep on working on the data. Let's say there is an e-commerce application. When you take a e-commerce application, you'll have a structure data, right? Like order ID, product name, price, delivery date. And uh structured data helps track purchases and inventory. Whereas unstructured data, under unstructured data, you'll have customer reviews, product images, social media mentions, chat support, interactions. These all comes under unstructured data. And unstructured data will have reviews and comments which helps analyze customer sentiment and improve products. And you can see when always when you are trying to make a purchase on an e-commerce website, you look at the reviews. Out of those reviews, you find the rating also for X product. 2,000 reviews has been given. And you'll see the rating like 4.4. That is a rating overall rating for the product by those many customers and you keep reading the comments and uh make a decision whether to buy or in banking also fraud detection will happen. Banking and product detection related. What kind of you'll have transaction history which is in the form of structured data like account number, amount, location and time. When it comes to unstructured data, customer emails, phone calls, and social media complaints. Structured data is used for tracking the balances and transactions. Whereas like customer complaints about fraud on social media helps detect suspicious activity and improve security. Right. So let's take it forward. I was talking about data science emerges from the combination of subject expertise and scientific methodologies and technology. Here you can see domain expertise and scientific methods. You you have a large amount of data in hand and with the help of mathematical and statistical models you'll model the data then perform some kind of analysis. After performing some kind of mathematical modeling over the data, we perform the analysis then apply some scientific tools and methods and use programming languages, data processing tools with the help of operating system, the predefined library support with and uh get into the data science and find the insights of the data. Right? By this data science is a combination of subject expertise, scientific methodologies and technology. With all the combination we are able to achieve the data space. You can see an application of data science in healthcare where variable devices use data science to analyze the data from their biometric sensors. What is happening? You have a variable device. the data from the variable device. For example, every day you are going for a workout and burning some X calories out of your workout and you're also performing some 10,000 to 12,000 steps every day. Whatever the data is collected on the variable device is transferred. Biometric data transfer is happening through the IoT gateway. Internet of things gateway. As variable device is a IoT device. It is controlled by internet. So the data which is available on the variable device is transferred through the IoT gateway. If you're using iWatch, I watch gateway will be used and data is transferred to the servers and uh I watch related the enterprise in will be there all the required data is being collected from millions of users. the collected data over the collected data they perform the data analytics like let's say for a week and for a month you'll get a dashboard on the engagement window when you open the application you'll get the dashboard with a lot of clear visuals workout related visuals calorie burnage visuals your uh number of footsteps All the readers you'll be getting right by this what you came to know the variable device through the biometric data transfer the data is being transferred to the IoT gateway and through the IoT gateway again the data is getting transferred to the servers of the enterprise infrastructure from where your data is pulled and performed data analytics here All users data is getting stored from which only your data is being pulled and perform some kind of analytics and uh performed analytical insights are being displayed onto the engagement dashboard in the form of visuals and when you open the application related to the variable device on your smartphone you'll be able to see all these visuals and you'll be keeping informed about the last one week workout details and everything. So that's how the variable device will work where you get all the information on the dashboard. Hey folks, are you following? Is the place comfortable? I would like to know at this stage what are the recent searches has been done. Most of the hits has been done. Those are kept as a reference and it started referring to those suggestions due to which your search becomes very fast and realtime analytics is made possible to buy modern and advanced infrastructure tools and technologies. With the help of the data science being used in the search engines, you are able to perform the search with the help of suggestions, right? And applications of data science and finance. Finance is talking about this is a very good example which will make you clearly understand everything. Let's say using data science a loan manager can easily access and sift through a application's financial details. As a user, I'm applying for a loan and you have a loan application portal. What loan application portal is asking you? It will ask you some details like your name, your salary to which bank is getting credited. Right? Also whether you are residing in a rented house or own house, marital status, recent salary, hike, company name, many details it ask details. After taking all the details, it will transfer the entire collected data onto the servers. Okay, will transfer enter data collected on the portal to the servers. You have a enterprise infrastructure here which performs the data analytics over your data and uh transfers the data to the engagement dashboard and engagement dashboard will generate your credit history, credit report, approved amount and the risk all the details the rate of interest the amount you are eligible your civil score all these details will be displayed to you on the dashboard and also you'll keep informed whether you are eligible for the loan or not eligible. That is how the loan online loan application process. Earlier taking loan was a very tedious process. You need to call the bank exeutor. is to collect the pay slips lot of other documents and move to the bank process your application there was a lot of waiting period but all these are leveraged here a serno that is the beauty behind the data science next data science process what is the process involved in the data data science involves the Problem definition. What what actually you want to do is a problem definition. Later over which data you need to collect the data. After collecting the data see without data you can do nothing. So what kind of problem statement you are looking for and on which data? Okay. After collecting the data data may have lot of missing values. Missing values like might be phone number missing might be age of the employee missing somewhere salary is also missing right. So data cleaning and exploration should happen to handle the missing values. Then feature engineering should happen. What feature you are targeting? Model building and training. Then model evaluation. Model deployment. These are the different steps involved in data science process here. Let's look into the every detail step clearly. Problem definition clearly define the goal or question to be addressed through data analysis forming the foundation for the subsequent steps. That's what I was telling you right what problem you want to address is all about problem definition data collection gather relevant data sets or information sources necessary to address the defined problem. So you need to have data in hand. So collecting the data so that whatever the problem you are trying to address the relevant data you need to collect and uh keep it handy so that you can perform various other steps required for performing the data sets. The next step is all about data cleaning and exploration. What does data cleaning and exploration will do? It will pre-process the data by handling the missing values, outliers and other inconsistencies. What does this outlier mean? Just now I told you what is a missing value, right? Can anybody tell what is an outlier? Outlier is a very simple word. For example, if you take a class of students, most of the people who take exam might be scoring might be scoring marks between say if it is out of 100 marks out of 100 marks and there are 60 people most of the people might be scoring around 70 to 90 right and you can see few people scoring Five marks, 10 marks, right? While performing analysis, you don't consider these marks because we consider them as outliers. Consider them as outliers which is completely moving from the actual data. Outlier is a data point which is moving completely from the actual data. And when you consider these outliers, your analysis may go wrong. Your predictions may go wrong. So to avoid that to improve the accuracy, you'll ignore the outliers. Hope you're getting fine. Next, feature engineering. Feature engineering is all about create or transform new features to enhance data sets information and improve model performance. If you want to perform analytics over the data, sometimes you may need new features. So for creating new features over the existing data and adding it to the data set we go for feature engineering as you need large amounts of data to perform analytics and some kind of feature generation should happen. So you'll get to know more detail while discussing the feature engineering topic and model building and train developing develop a predictive or descriptive model using machine learning algorithms and train it on a prepared data set. We need to train the model on the data set. So develop a predictive or descriptive and train it on the prepared data set. So we train the model most of the time we train 70 to 80% with and uh test with the 20 to 30% of the data. So the ratio goes that way. Then model evaluation and deployment. We evaluate, optimize, and fine-tune the model for peak Then deploy it into a production environment for real use. So evaluation, optimization and finetuning of the model will happen. And I say interpreter, right? interpreted means for example you take any other language for I am taking some five lines of code I'm taking five lines of code and here I have error at line number two and also I have error at line number four if it is a Java or C# code C++ code also if I give this code let's say this is a test dot Java when I give it to the compiler it will compile all the code on one go and give you the list of errors right list of errors saying that there is an error at line number two and there is an error at line number four when you consider the same code to be a Python code same code to be a Python code And let's say it is a test. py. When I give this code to an interpreter, it interprets line by line. When there is an error at line number two, it stops compiling at uh interpreting at line number two and it gives you an error. Unless and until you correct this error, it will not take you to the next line. Okay. Let's say I have corrected this error and again submitted to the interpreter. It interprets all the three lines and at the fourth line it has found the error and it again prompts you the error. When you correct this error it will take you the next lines. So line by line interpretation will happen in Python whereas in Java the compilation will happen the entire code will be compiled on one go. And Python is a high level language. How we can say it's a high level high level language? Whatever the instructions you write is closely resembling the English language. Hence I say it is a highle language that supports object-oriented programming. You might be aware of object-oriented programming like uh the object-oriented programming principles, abstraction, encapsulation, inheritance, polymorphism, right? So, Python is supporting all the object- oriented programming principles. Hence, we say it is a objective Ease of use and simple syntax. How you can say ease of use and simple syntax? This is very important. Let's go and open online JDP compiler right. If you want to print hello world using C language you can see I'm selecting C language. So only to print hello world we need to write these many lines of code. How much? 1 2 3 4 5 six lines of code. when you run still a minute when you run you'll get hello world can you see but when it comes to Java also you can see 1 2 3 4 5 six lines of code again I'm running we get the hello world this is related in char but this beauty of Python you'll have only one line of code to print hello world just you run it you get it that is the beauty of Python with one line of code you are able to print hello world when it comes to the other languages it take it is taking more lines of code right hence I say it is simple syntax and ease of use very handy to program. Earlier most of the universities to used to have C as a starting programming languages but now most of the universities has revised their syllabus and came up with Python as a first programming language. That is the beauty of Python scalability when compared to R. Python applications are more scalable. It can be given to large number of users for accessibility when compared to R. Hence we prefer Python as a best suitable language for data science. Availability of wide variety of data science libraries and patterns. We have different libraries like numpy, pandas, skypi, seabon, mro lip and many more which supports data science related processes. Hence I say Python is best suitable as it has wide range of libraries and packages. It is compatible with all major operating systems. Of course Python you can write Python code on Mac can write Python code on Linux you can write Python code on Windows many more. creation of new data science libraries daily by a vast number of online user communities. There are coming up new libraries which supports Python and using those libraries you can perform data science. You also have a powerful data visualization libraries in Python. Hence we recommend Python as a best suitable language for data science. Have you got the clarity here? After this discussion participants even Nin Raj I hope your doubt got clarified. Great. Great. Next, it's right time for us to look into the packages for data science. Numpai. Numpai is a Python's library for data scientific computing that supports large multi-dimensional arrays and matrices and includes a comprehensive mathematical library. What is that number is used for? Please try to understand if required document them very simple oneline thing what is that numpy is used for it's a python library used for scientific computing and that supports large multi-dimensional arrays and matrices I can tell you a simple thing for example you want to perform matrix multiplication if you want To perform matrix multiplication 2x2 matrix through programming, it may take 20 lines of code in Java. If you want to perform 3x3 matrix multiplication, it may take 30 lines of code in Java. But when it comes to Python, it hardly takes one line of code with the help of numpy. That is a beauty. So you need not write any hard coding for performing matrix multiplication. You use the required function and you'll get the desired result. That's it. That's the beauty of numpy. Use numpy to perform complex mathematical operations on large data sets such as linear algebra calculations, statistical analysis, and 4year transformation. You can see here financial analysts use a numpy for quantitative analysis such as calculating the mean and uh volatility of stocks to invest to make investment related decisions. So that's how numpy is useful, So what is numpy? Numpai is a Python library for scientific computing and for working with large data sets to create large multi-dimension arrays we go for and pandas. Pandas is a library for efficient storage and manipulation of structured data such as time series and tables. Here e-commerce applications use pandas to analyze customer purchase stream to recommend products and personalized shopping experience. Pandas contains data frame which will help you in storing the data. in the form of a structure structure format let's say there are as I told you right recommendation of if you are trying to purchase a smartphone immediately it will give you some suggestions recommendations people who bought a smartphone also bought the screen guard, a pouch and earpods, many more power bank. it will analyze the data and it will analyze customer purchase history to recommend the products and personalize the shopping experience. Researchers use the pandas to manage and analyze large data sets of health records to identify trends in the disease outbreaks. When it comes to the health related issues, for example, some ex person has is might be suffering with heart related disease. Researchers will use these reports and uh relate or perform some kind of analytics with the some other thousands of records and we'll identify the trend and come out with the discussion. A person with the age age range between 35 to 45 only 2% of the people are getting these kind of diseases and there is a chance that 20% of the people will get at this age. So this kind of analysis can be performed over the data using pandas. Skype skyp stands for scientific python is an opensource library built on top of numpy and is used for implementing scientific formulas pval np and so on. It is tailored for scientific and engineering applications such as weather forecasting and drug discovery. So engineers use SkyI to solve mathematical complex mathematical problems in structural engineering such as stress and strain analysis in materials. Okay. So most of the time Skype is used for structural engineering and weather forecasting, drug discovery Stats models. Stats models in Python module that provides classes and functions for estimating many different statistical models and conducting statistical data exploration. Like we have a couple of examples here. Market researchers use stats models to perform regression analysis to understand how different factors such as advertising spend affect sales. Right? Advertising spend is affecting sales. It's like say there is a X product X and uh I want to improve it sales. So to improve the product X sales I want to invest some Y amount in the market for marketing and advertising the data and I'll analyze the trends before mark before advertising and after advertising when I perform modeling on the data and pull the data related to sales before adver advertising and after advertising. Yes, I'll be able to understand whether the it the advertising amount spent is affecting the sales or not. And also product analyst will use stat models to evaluate the impact of public policies on social media and economic outcomes through statistical testing and analysis. Skyitlan. Skyitlan is a widely used open-source machine learning library for Python known for its simplicity, ease of use and versatility in handling various machine learning tasks. What can be the various machine learning tasks? You can see here there are few examples. It identifies objects in images for autonomous vehicle and facial recognition system. You can see autonomous car vehicles driverless cars. So driverless cars are using sky skyit learn to identify the objects and uh accordingly make a autonomous movement in the vehicle. It is also helping to detect fraudulent transactions in banking and e-commerce platforms. Recently I was trying to perform a transaction of a big amount on some platform. My transaction was denied and immediately I got a call from the bank and it is a IVR call interactive voice response call. It was asking the recent transaction has been happened on your card with the so and so date so and so time on so and so platform to confirm say yes s and I said yes s later I was able to perform the and it is detecting the fraudling transaction. It might be having my spending history and uh none of the spending history is matched with the new transaction I'm making that two big amount. So immediately I got a alert from the bank. So internally it is performing the skyit learn to detect the fraud transactions in banking and e-commerce applications. It is also analyzing the customer reviews for sentimental classification in marketing and social media analysis. You can see for a product you'll get a rating automatically and uh you can click on the different tabs to understand the sentiment analysis whether positive, negative or neutral. If it is negative what way they are not unhappy. If it is positive, what are the best uh things related to a product? And if it is neutral, obviously we are not worried about the neutral comments, So, Skyate plan will help us to analyze the customer reviews for sentiment classification in marketing and social media analysis. Map plot li Map plot lab library is a comprehensive tool for building static animated and interactive visualization. See always remember instead of displaying the data raw data or a process data visual speak much louder right when a cricketer is walking onto the ground lot of statistics gets displayed about his recent matches. What are the recent matches he has played? What is his performance? If he's a batsman, how many runs he has made in the recent matches? Every detail will be displayed on the screen. So, visuals are much louder than the actual process data. Hence, visualizing data is most important. Even a lay man will be able to understand when he is presented with a visual. And math plot lab is helping us to create such static animated and interactive visuals including the line plots, scatter plots, bar charts, histograms, pie charts and many more. And matt plot lip is a open source library can be used and we discuss mapplot lip in detail in the upcomation library but this is built on top of macro. The reason it provides a rich visuals compared to macro clip. It provides highle interfaces for creating attractive and informative statistical graphics like histograms, box plots, violet plots, heat maps and error bars. It also simplifies the process of creating aesthetically pleasing. Okay. and informative plots especially for statistical and categorical data. Fine. That's all about the C1. Next, plotly. Plotlay is a again Python library only for creating interactive publication quality graphs and visualizations. It is suitable for web- based applications. It enables Python users to create interactive and customizable visualizations for data analysis. what is that interactive and customizable visualization it is helping us to create. Meanwhile various plot types it supports various plot types such as line plots, scatter plots and histograms which enhance data exploration. Types of plots with examples. Now types of plots we call this is a line plot. A line plot displays the data points connected by straight lines. Can you see these are the data points number one data point two three and four. Four data points are connected. Often used to visualize trends or relationship between two variables over time or continuous or other continuous intervals. A land a line chart visualizes stock prices over time tracking trends and marketing investment decisions based on historical data. You'll be able to access the PBT in your materials if required. I'll share you. Not a From my side also I'll share. and it displays temperature variations throughout the year to analyze seasonal patterns and plan agricultural activities effectively. So line plot is will give you some clear understanding between the two variables and variable trends like uh for example let's say a product X price in 2010 and a product the same product price in 2020 and 2026 year wise price you want to map obviously it will be a line chart increasing trend, right? It will be like a line like this. Oh, sorry. It will be a line like this. Straight line. It keeps on increasing. So, as the year increases, price is increasing. That would be the trend you can understand. So, relationship between the year and price. Next marker plots. A marker plots display data points with markers. You can see marker circle as a marker, right? Useful for scatter plots and visualizing individual data observations. Marker plot is used to display individual data points on map such as marking specific locations for a survey. It is employed to plot stock prices over time with the markers indicating specific events like buy or sell signals. at this point selling is happen at this point buying is happen that way it is going to indicate you the marker plots. Scatter plot. Scatter plot is all about the collection of data points plotted on two axis both horizontal and vertical. You can see lot of data points scattered here. What does this data point indicate here? See here it is indicating the data point related to x-axis is somewhere close to right close to two and here close to 4.5 right 4.5. 2, 4.5 would be the data point value. That way x-axis and y-axis plotting will happen. The scatter plot analyzes relationship between two variables like comparing height and weight in a population and it helps us to visualize the data cluster such as grouping students based on exam scores. So most of the data points are here right if you can understand most of the data points are residing here and also here right so that way we'll when you can when you collect let's say there is a there is an apartment with 500 plats There is an apartment with 500 flips. In a apartment with 500 flats, we are collecting the income of every house. Income versus age. Income versus age cap capture income and age. and uh you keep uh plotting the plot then you'll get to know what age of people or more with a particular income group. So that kind of analysis can be performed with the help of scatter plots and the types of plots. Area plots we have area plot represents data with shaded area useful for showing cumulative proportions cumulative totals or proportions over time. And you can see here A, B and C, An area plot visualizes cumulative data changes over time. such as tracking total sales revenue over successive quarters. It illustrates the distribution of data categories different categories contributions to a whole like displaying market share evolution. Right, that's all about area plot. We'll discuss area plot through an example as well. Don't worry. And types of plots. Bar plot. Bar plots are rectangular graphs that show identical uh vertical and horizontal data comparisons based on another axis. A bar plot compares sales of different products over month. Sales of different products over a month. Say for example, I want I'm taking some five month data analysis. On month one products X got sold only 25 units later close to 50 units. Later later 50 above units later 75 units. Later in on fifth month altogether 200 units got sold. So that way you can relate the bar plots. It also displays student grades in different subjects. Grid plots. Grid plot assist chart viewers in determining what value an unlabelled data point represents. Grid plots enable sideby-side comparison of multiple plots enhancing visual analysis. So you can see it's it's almost similar to the line plot but you have a grid in between right? Can you see a grid? They enhance presentation clarity by organizing complex information systematically. Likewise, histogram. Histogram visually displays the distribution of data set by dividing value into bins and representing the frequency of each bin and bars. Here bins we have we divide the data into multiple bins. Multiple bars nothing but okay. so that we can categorize the data in a very clear way. Histograms visualize the distribution of numerical data like income levels or exam scores. They help make inferences about data characteristics and underlying patterns guiding decision making processes. So you have different bins and uh by this what you came to know your data is broken down into multiple pieces and uh grouping them as a bins so that further analysis over the data will become very easy for us. Pie chart. Pie charts are circular graphs in which data are plotted within the components or segments of the pi. See data is plotted here. Pip plot shows the proportions of a whole like market shares or survey responses. Out of survey conducted here few people 30.7% of the people got registered for Ruby and 23.56 people uh sorry 26.3% people were registered on Java and so and so they simplify complex data by representing it in an easily understandable format and most importantly examples of all these plots are provided in data visualization lesson accompanied by detailed explanations and Python code. So need not worry if you have understood to whatever extent you like feel comfortable and the remaining part you'll get more clarity while discussing the data visualization lesson. Fine. Yeah. Key takeaways we have seen data science involves the analysis and interpretation of data to generate actionable insights. Numpy and we call it as a numerical Python is a opensource library. Predominantly used when working with arrays. Seabbone is a data visualization library in Python that is built on top of M plot lab and Python is preferred programming language for data science projects across industries. Okay, hope we can proceed. Let's look into the introduction to numpy. talks about fundamentals of numpy, advantages of numpy, numpy installation and importing and uh numpy array object and creating numpy arrays. What is numpy? As I have been telling you number stands for numerical Python is a free and open-source library that is mostly used for mathematical operations in scientific and engineering applications. It is a Python library used for working with arrays. It consists of multi-dimensional array of objects. multi-dimensional like two-dimensional threedimensional arrays we have and a collection of functions for manipulating them. So you need not write any code for that only all the functions are available. You are only using it for getting your work done. It conducts mathematical and logical operations on arrays. What does it do? operations on arrays. And the the array object in numpy is called nd in dimensional array. What is that? We call a numpy array as a n dimensional array. What are the advantages? It provides an array object that is faster than the traditional Python list. and also it supports functions. Arrays are frequently used in data science and numpy arrays are stored in one continuous place in memory unlike list and they are stored in one continuous place. on continuous place continuous memory locations numpy installation generally you if if you're trying to import numpy let's say introduction to numpy you change the code to markdown and run it you'll have some text here then I would like Write a simple print statement to check the environment printed. Welcome to numpy. So run it. Are you getting? Yes. Our platform is ready and we can port for numpy. If you're importing numpy, import numpy and uh after writing this statement if you're running you're getting nothing indicates numpy is imported. If you get a error you need to by default actually numpy is available. If you want to install numpy, you need to say pip install numpy. If you want to install but here we succeeded in importing numpy. So we are not running install numpy. So numpy array a numpy nd array n in dimension array object can be created using array function. What is that array function? So here we go and create a simple numpy array. I import creating a numpy array import as np np is the alias name I'm giving to numpy so that we refer to numpy using np In this context, numpy is no more existing only np is existing with the help of alias name only you'll be able to identify right and I'm I'm taking a variable a r ar ar ar ar ar ar ar ar ar ar ar ar ar array is equal to np is a numpy alias name over which I'm calling the function array array is a function related to the numpy to which I'm providing the list of values. Okay, numpy is an array to which I'm providing list of values. Array takes a list of values and converts it to a numpy array and will be stored into the array. when you're trying to print array, you'll be able to print array. And uh when you try to get its type, type is a function which takes the variable arr and gives what type of data is associated with the array. So I'm running the code. You can see list of values is converted to a numpy array. which is a n dimension array. Can you see when I'm trying to get it type it's a n dimension array I want you to put your hands onto the keyboard complete the demo you try taking different values and uh post me the output this is the first activity you need to do which is a mandate to understand whether you are ready with the environment and able to perform the hands-on activity As it is simple piece of code, if the lines of code is more, I would recom I would be posting you the code which will really help you in taking the code from the chat and uh executing in the and happy to see Wuni Rao and Sanjuaki has done with this only four participants. has posted so far. Waiting for many more responses. I shall wait few more minutes if required. But make yourself comfortable to the first demo which boosts your confidence and energy towards learning. If you have not understood you, please let me know. I'll make you comfortably. Yes, I can see Sumit Ranjan also has done very good. Only five participants so far. Aila Sultana Ashish Bam James Sumit Ranjan Sumitan has done sorry how you doing please respond as in one more minute we'll go ahead with the next demo Next demo is all about creating multi-dimensional. Okay, good. So creating a zero dimensional array. I'm taking a variable array zero is equal to np dot array and I'm giving a single value to it. I would like to print z array is and array zero. Let's think how it looks like a single element zero dimension array. Yes, zero dimension array will have only one element. You have good good observation. Next, I would like to create a 1D array. Array 1 is equal to NP dot array of 4. print 1d now I'm running so one dimension array just now we created it is also one dimension array array 2 is equal to np dot array I would like to Take one and uh 1 2 1 but I would like to take Can anybody say me quickly what is it? I'm selecting it. I want you to tell me what is it. Here it is a list right? And what we call it as are you getting three dimension array is a combination two two dimensional arrays. Here you can see zero dimension array is having only one element. One dimension is array is having a list of values. Two dimensional array is having list of lists you have two square brackets right. So two dimension array is having two square brackets. Three dimension array is having three square brackets and three dimension array is having two two dimensional arrays. Can you see this is one two dimension array and here is another two dimension array. So combination of two two dimension arrays is giving you a three dimension. I would like to share you the code initially as many participants may not be able to catch up while hearing typing the things and uh getting the demo done. So to make you comfortable not for everyone who is not very much comfortable I can help you take the code from the chat window and run the code initially feel comfortable later over a period of time by spending enough time ample of time by performing the hands-on activities you can make yourself comfortable post learning post session fine so even for this also I'm expecting the responses from you I shall wait here we are done with the introduction to numpy we shall go for the next subtopic in numpy So instead of always moving, I'm pulling two lessons, two sub lessons into the practice labs. And uh here are the attributes and functions in Python. Hope we are good to move for the next. When it comes to attributes and functions in Python, attributes of numpy array and the functions of numpy array we discussed. Attributes you can see few common attributes of numpy are listed here. First one is array ending. Then shape, size, data type, item size data. These are the six attributes we have. One after the other we'll see through an example. Now what does it give you? Shape of the array. A rer dot over the array. I'm calling shape. It will give me the shape of array. A r. What is the shape? We had a discussion in the earlier example. You are right. It is two rows, three columns. Two rows, three columns. Fine. Next. You want me to zoom it? Yeah. Is that okay? Now, comparatively, it's fine, right? No. I want to look at the attribute size. What does size of array gives you? Any guess? Size of array simple number of elements. How many number of elements do we have here? That's it. It gives me number of elements as six. So before we proceed, I'm running this. Can you see number of dimensions is two. Shape of array is two rows three columns. Size of array number of elements is nothing but six. We have six elements. Next I say array stores elements of type array type. What type of elements does it store? Very good. You're right person. It is a integer type of Can you see integer 64? How much how many bytes of data is allocated for one element of an array? Length of one array element in bytes array dot item size. So every these are the array items specific item size I'm trying to get so that that will be the memory allocated for every element in the array. What is its size? It is 8 bytes. It is eight bytes here. And finally I would like to print it arrays r data. What does it give you? It gives you the memory location where the data is stored. Fine. I want you to perform this demo. Okay. The one we are going ahead attributes we have already discussed. So we shall look into the so the all explanation I have given you later we'll move forward to the numpy array functions we have transpose reshape and flatten what does reshape do it is used to reshape I mean giving a new shape to the current elements of the array let's say I want to convert But converting a 1D array to a 2D array. array equal to np dot array of new array I'm saying is equal to over the existing array I call reshape what is that over the existing array I call the reshape when I call the reshape what happens It will reshape to the given size like uh 4a 3 I say. Can can anybody tell me what will be the shape of this array? Okay. So I can see the responses. Very good. Very good. four rows, three columns. Let's check whether it is reshaped or not. I'm trying to print the array and also I'm trying to print the new array. Right? Can you see this is the given array. This is the converted directory. How many rows? Four rows and three first column, second column and third column. So reshape function is converting the required shape. So likewise let us look at the flatten. What does flatten doing? It returns a copy of the array flatten into one dimensional. If you have a three-dimensional array, so to save time I'm taking this threedimensional array array equal to or array 3D equal to array 3D equal to np array also I'm taking the 3D array and I say flatten is equal to array 3D dot simple I call flatten flatten is a function which will convert any dimension to one-dimensional. So you can print them. Can you see a 3D array has been converted to so I'm creating the document in a such a way it will be really helpful to you in next learning after is did I give you the code earlier shape? No. Yes, I given. So I would like to give you the code related to flatten as well. Again the flatten array can be reshaped. How you can reshape? You can call flatten array dot reshape. Whatever the shape you want, you can reshape. Okay. If you're looking for if you're looking for converting it to the original shape, what you can do? I can I'll do that print array dot reshape of. So in the earlier session we have been discussing numpy arrays how to create an array how to create a single dimension array multi-dimensional array and uh two dimensional threedimensional array and how to reshape an array. How to plat an array? How to transpose an array. So these were the different things we have been discussing in the earlier session. Right? I'd like to rename the file day two numpy and pandas numpy arithmetic and statistical functions. We have arithmetic operations using like addition, subtraction, multiplication, division and power off. In addition to that we also have statistical functions of numpy like for calculating median, mean, standard deviation and variance in the Also calculating percentiles and uh string functions in number arithmetic operations using number we have two arrays I would like to perform addition okay so let's take this example arithmetic operations. Please let me know if I'm not comfortable with the pace arithmetic operations in I'm taking two arrays numpy as np and a equal to np dot array of 10 20 30 and B equal to NP dot array of 30 30 201 then I would like to perform addition. So in order to perform addition what is that we are using here add is the function we are using right. So result equal to np dot add I'm passing array a and b to the add function so that it will perform addition of both the arrays and here respective elements will be added index related see zer index element will be added to the zeroth index first index element will be added to the first index likewise and finally I would like to print the result you can See what is the reason power I missed R. Can you see what is happening? A to the^ of b 2 ^ of 2 ^ of 3 is 8. 2 ^ 4 is 16. Likewise you're getting 2 to the^ of all these. Can you see now? Yeah, sorry. Is that fine now? So we were performing power power operation for all the elements. So a ^ of b 2 ^ of 2 2 ^ of 3 2 ^ of 4 likewise 2 ^ of 6 2 ^ of 6 is 64. Final objective is to make you comfortable. If the pace is not comfortable, if you're not able to switch between the windows, understand the explanation. I recommend you to please respond immediately which is really essential. Now it's time for us to discuss about statistical functions. What is the statistical functions? Oh no, it may not work. It may not work. It it say for example I'm taking here an equal size. Is it working? No. So equal number of elements should be there for both the rays. Are you comfortable? Is that okay? Let's look into the statistical functions in numpy. So I would like to calculate mean calculating, median, mean, standard deviation and variance in array. So median array equal to np dot array. I'm taking the elements 4 3 into 10 1 0 5 8 So what what dimensional array it is? Can anybody tell me what is the dimension of array? What is the dimension of array? NP do median of what will be the median? Can anybody tell me what will be the median? twodimensional array Ashto. It is a two-dimensional array. It's not a threedimensional array. Very good. Sectan and uh Syani has given the right answer. Yeah, median is a middle value, right? Median is a middle value. But you need to arrange them in order ascending order, right? after arranging them in ascending order getting the middle value is the median. So shall we arrange them in ascending order? Let's arrange them here. See I'll arrange them in ascending order. It is starting at zero 1 then 2 3 4 5 then 8 then 10 then 24. How many elements we have got? 1 2 3 4 5 6 7 8 9. So I'm trying to get the middle element. I'm reming right four elements. I'm reming right four elements. I'm reming left four elements. And here you'll be getting a middle value which is a four. So four will be the med. Have you got the concept? Four will be the median middle value. I take I would like to print this in the printer medium. Then I would like to calculate mean. For calculating mean, mean is nothing but an average. NP dot mean of array. I would like to print it. What are the calculated value? I'd like to print it. What does it mean? Sum of all these values by number of values is right. Likewise, standard deviation will tell NP standard deviation of This is standard English and also variance. I would like to find variance. Print np variance is a function which will take array values and give you the variance value. That is the variance getting so you might be knowing what is the median middle value what is the mean average let's calculate the average also now I'm opening the calculator and See, I'm adding 4 + 3 + 2 + 10 + 1 + 0 is not required. + 5 + 8 + 24 = 7. So, divided by 9 elements, right? 9. How much is the value? Are you getting the same value here? 6.33. Same value we are getting, right? That is the mean. What is the standard deviation? Standard deviation all over the how much each and every value is deviated from the actual mean. See four is divided. What is the deviated? 2.33. 3 is deviated 3.33. 2 is divided 4.33. Likewise these deviations it is going to calculate that standard deviation. And you can see the formula for standard deviation. Formula for standard deviation goes this way. Can you see square root of the variance is nothing but standard deviation. Sigma of X I minus mu. X I is nothing but in this context X I is nothing but in this context every individual value it would be four it would be three it would be two so it's calculating 4 minus 6.33 plus 3 - 6.33 + 2 - 6.33 like every value it is subtracted and 4^² by n under root is nothing but a standard deviation you can calculate you'll get the same value next variancy nind manage Variance is square of square of 6.99. See calcium taking 6 9 9 9 44 42 square are you getting 48.22 two. So standard deviation sphere is nothing but variance man. Are you comfortable now? Now let's go and calculate percentiles. Percentics calculating percent here np dot percentile of array comma 50. What is the 50th percentile of an array? This is an array. When you arrange in an order, what will be the 15th percentile of an 15th percentile of an array is nothing but median. Always remember there are three quartiles. If there is a data point, the data points are divided into the data points are divided into two quartiles with the help of Q2. two quarters into with the help of Q to we call it as median the left half is divided into two halves again with the help of Q1 quartile this side will be the 50% of the data and the right side will be the 50% of okay and Q1 is dividing this 50% into 25% 25%. Likewise, you also have a Q3 Q3 which is dividing the remaining up to 25 percentile and 25%. So by this what you came to know there are two quartiles there are three quartiles sorry what are they Q1 Q2 and Q3 these three quartiles are dividing your data into four equal parts and uh Q2 is called as a median Q3 minus Q1 is called as interquartile range. What is that? Interqu quartile range. In this interquartile range 50% of the data will sit. In this interquartile range 50% of the data will be residing. You can see 25 50% right. So 50% of the data will be residing. That is understand. So percentiles the quartiles will play into the come into the picture. You have three quartiles Q1, Q2 and Q3. Q2 is called as a median which is dividing the entire data into the two equal parts and Q1 is dividing the left half into the two equal parts. Q3 is dividing right side into two equal parts. Q2 is called as a median. Q3 minus Q1 is called as inter quartile range where 50% of the data will be resided. That is our understanding about the percentiles. Any doubts here? If you want to generate the sales report like I want to find out 50th percentile salary is the amount at which half of the employees earn more than and half than less. When you get 50 50th percentile of the salary, you you'll get get into a conclusion 50% of the employees will earn less than the this salary and remaining 50% will earn greater than that. If you test some score and is it is in 90th percentile what you say you score better than 90% of the students when a particular person score is 90 percentile you can say you you scored that's how we differentiate in the real life so I have taken two arrays they are string arrays. I would like to perform the concatenation. If you want to perform a concatenation, you have a function called as cap. Under the numpy you have a function called a scare under which you have a add function which takes two arrays and uh add them or concatenate the respective index elements and will give you the result. What is that? Hello will be added to welcome bold will be added to the learners and you'll be getting the respective result. Can you see what is that we are doing here? numpy contains a character which contains a function add taking array x and array y and printing is the result. So if you want to generate an email you want to generate an email with the combination of first name and second name or first name and last name combined with at the right company name. This would be useful or if you want to create a URL or if you want to like label the data adding the prefix or suffix to the existing data this will be useful. If you want to add some spaces you can add here. Can you see I'm adding some spaces here and later I'm running the code. You can see there is a lot of comfort. Hello, welcome right so space is I like to replace how do you replace I'm saying np dot car dotreplace over the car class you have a replace function which takes the string and uh replace in the string hello with hike Hello with I would like to store the replaced string the new string into the set so that I can print the new string. Can you see is the old string where we have used car dotreplace of string. We are passing the original string and in the original string I'm replacing hello with five. So replace is taking three parameters string parameter the old string and a new string. The entire text with the old string and a new string. So that the old string will be replaced with a new string. And the the updated string is stored into Z zed and I'm printing that zed. Can you see that's how it's functioning and uh next is all about I would like to discuss so replace is all about in word you'll find and replace right control F you'll find a particular text and try to replace with a new text. Here is a very interesting example related to replace. I would like to discuss one more example. I have names and uh updated I would like to is equal to np dot par dot replace so I would like to replace all the names all the names contains mister right I would like to replace the mister with doctor now it's all about indexing. Indexing you have accessing elements in a one dimension array, two dimension array, three dimension array and native indexing also we have numpy array is providing you Indexing allows you to access the array elements using a index value beginning from zero. We have three arrays, threedimensional array. So let me take this array to save some time. I'm taking this We have three dimension. One dimension, two dimension, three dimension. Now I would like to access I would like to access and I would like to print the value of index three I would like to print the value of index three so print array_1 So can anybody tell me what is the value at index 3? That is really important. Your interaction is makes more value. I have a three dimension arrayed one dimension two dimension three dimension and we are working on one dimension array. Yes, you're right. I got a few responses from Mani and UA. See index is starting at zero zero 1 2 and three. At third index you have the value four. Right? Anything? And also I would like to perform addition of index 0 and one. How do you perform index 0 and one primitive? Array 1D of zero plus array 1D of 1. So what does it give you? Zero index plus first index. It is a right. So I recommend you to please run the demo. So add function. No, we cannot use add function because add function is used to add two different arrays. But here we are adding the same array different index values and we perform it using indexing only next to two arrays. third element of the first row of a 2D array. What is that? Print the third element of first row of 2D array. Third element of first row. What is the third element of first row? This is the first row and this is second row. Right? What is the third element? This is the third element. How do you print? How do you print of the first row? ad worked like how did you use it? Oh, let me see the response. Okay. Okay. Of course. Yes, ad will work this way. No problem. No problem. Good. Good. Happy to see your response. That's really a interesting learning and can anybody tell me how to get second element of second row. Second element of second row second element of second row. In the second row, say this is a zero index. This is the first index. And again, it's index is starting at 0 and one, right? 1, one, right? Yes. Yes, you're right. And uh please participants try to I'm trying to take you through a very simple and comfortable learning journey. And if you asking me to very slow down the learning, we have a lot of things to be catch up that would really complicate our learning. at a very basic beginning stage. If your if your learning is at a very snail pace that would really impact your the whole These are very simpler things. I recommend you to stay tuned with the see 1 comma 1. What is that 1 comma 1? five first row this zero index row. This is the first index row and in the first index row I'm targeting first index and I'm getting five. mechanism that is how I'm accessing the two dimensional elements and how to print that element. array reading X corresponds to X X element right. So first element and y is corresponding to y row y row is nothing but and z column first column what you getting what is that you getting 11 you getting right but this This is not the element you are trying for. What is it that you are trying for? First element of the second array. This is the second array. First element. Now can anybody tell me what could be the values I can mention here? 1 0 0. Ah, you're right. 1 0 and zero. You're right. Are you getting seven? Yeah, you're It's really interesting to see the response from you. the native indexing. Negative indexing as you know negative indexing is counted from the end of the array and in negative indexing system the last element will be the first element with an index of minus one and the second element will be the index with minus2 and so on. I want to print the fourth element of the 1D array using negative indexing. array 1D fourth element. How do you get the fourth element using indexing? Here we have 1 2 3 fourth element. Right? Fourth element I can say minus1 -2 and minus 3. Right? So when I sayus 3 I'll be getting - 3 can you see what does this indicate you have a two dimensional array in which I'm targeting second array what is that second array what is the second array This one this is the zeroth index array. This is the first index array. So I'm mentioning one in which minus one I'm seeing. So I would be expecting a minus 6 as output. Right? So am I getting - 6 as output? Can yes that's how it's functioning. Okay. Is that fine? Of course. One problem, multiple solutions. We have one problem, multiple solutions. Even this also will work. And I want to print The last element 3D array using native indexing array 3D 1A 1 minus Say let's see what is it going to print 1 - 1. See here you have three dimension array right one means this is a zero index array. This is a first index array. So it is referring to one in which you have one and minus one. Sorry in which you have this zero and this one. Right. Let me write this is a zero index and uh this one is a first index. It means you have again in the first index you have again zero index and What I said 1 minus one. So one is referring to this one. Next one is referring to this one and minus one is referring to the last minus one. Hence you'll be getting 12. Let's run and see whether we are getting the desired result. Are we getting Yes. Yes. Sorry. Sorry. Whatever you have mentioned coming to coming to Manshi and uh Sikki even now in Dutch. Yeah. Everyone is right. We are trying to get the elements from first index or seventh index. So what is the first index? First index is four. 7th index but seventh index is not inclusive. So we'll consider till sixth index. Or you can consider the other way. What is that? 7 - 1 is six. Right? All together you'll get six elements. 1 2 3 4 5 6 we'll get six elements we are getting the same six elements right that's how it is going to function slicing I would like to slice I'm taking a books is equal to np dot array of I'm taking some books physics Data science, maths, Python, Adobs, Java. and cloud. uh can take it here. This is a list of elements array of elements. I'm saying books of five column. What does this indicate? Starting from five, what is the fifth index? 0 1 2 3 4 5. So starting from five till end. If nothing is given at the end, it will consider till end. So from five till end it is going to print. Is that okay? Shall we continue slicing using step value? Hope everyone is aware of x = np array and I'm taking an array like 8 comma 7 6 5 4 3 2 1 then I would like to print x of 1 6 3 What does it mean? 1 to 6. Let's say index is starting from one here to six. What is the six index? 1 2 3 4 5 6. So 6 is not included. Right? Till here only from here till here. The step of three means 7. After that, what is the next step? 3 + 3 6 5 4 7 and four will be the output. What is that I'm expecting? Let's run and see what is that. Yes. So 7 and 4 step index at a step of three. Yeah, definitely it that would be really pre for me if you ask me to explain again. Yes, I'm ready. Always see starting at first index, right? This is zero index. This is first index, right? First index starting ending index is sixth index. So 1 2 3 4 5 6 but six is not inclusive Six is not inclusive. So I'm considering till sorry and at a step of three step of three means it is a first index right? First three 1 + 3 is four right? 1 2 3 4 What is the fourth index value? 7 and four will be the output. That's what you're getting here. Are you comfortable? Slicing to the Z equal to 11 22 33 Wed by 66 77 88 99 and I'm saying hey sorry one second I would like to change this window. Now I would like to print Z of 0 col 0 2 col what does it indicate? Zero index. What is zero index? This one. In the zeroth index starting at second index 0 1 2 second index right and indicate third index. What is indicate? It is targeting in 33. That's it. Once again you have two indexes. What is that in? It's a two dimen uh two dimension. This is the first dimension. Z index dimension. This is the first index dimension. So here I mention zeroth index dimension and I'm saying 2 col 3. In the zeroth index dimension I'm targeting second index 0 1 2 2 to 3 I'm saying 3 means 2 to 3 nothing but 3 - 2 1 only one that is 33. Now I want one second. Z is upper case here. you to help me in printing 77. I want you to print help me in printing 77. I'm waiting for your responses folks. Please go ahead. I want to print 77. I want your responses here. Okay. So, I'm right away putting the responses. You have shared me Z of 1 colon. What is it? You have sh 1 col 1 col and you'll be getting 77. Yeah, very good. Very good. Very very happy. Si waruni para now I'm explaining you once again. Please listen carefully. See this is a two dimension array and I'm considering this as zero index and it is I'm considering this one as a first index and here I'm saying while slicing I'm saying z of zero zero is referring to the whether it is referring to the first array or the second array and out of which I'm trying to slice right can you see this 2 3 this 2 3 if it is one it is applied here. If it is zero, it is applied here. Hence 2 3 you are getting 33. And when I say one, it is applying here. 1 2 nothing but 2 - 1 1 7 Is that clear? and all other participants are you comfortable please let me know actually for today's session if we go at a normal pace very normal pace we would have started the pandas part and completed almost 20 to 40% of the panda us understanding. So we are taking the session in a very very snail pace. The only reason I want to make you comfortable and uh we need to improve this pace. Please think of it, start working on it. No, it's all about slicing 3D array. I'm taking array 3D 1 2 3 4 5 6 then 7 8 9 8 10 11 12 It's a threedimensional array. What is it? A threedimensional array contains two two dimension arrays. In this context, print array of zero comma 1 cola 1 col what is it indicating you I'm targeting zero index. What is the zeroth index? This in which I'm targeting again first index. What is the first index? This one is the first index. Zerith index. In the zeroth index, we have zero index and first index. In the first index also, you have zeroth index and first. But I'm targeting zero index, first index and first index. So this is the first index in which I'm targeting first index. So 1 col means 5a 6. It will display me 5a 6. And I want you to tell me I want to print 11 12. What could be the slicing for three dimension array? What the what kind of notes see what are the handouts are there with you the slicing mi I will respond to you just let me clarify part's doubt para you have lot of handouts given to you right like uh slicing you can see you have a lot of things given right so you try to document then that's it that would be enough and indexing you have try to document only important on you try to document we're discussing slicing 3D and I want to display 11 and 12. Can anybody give me the inputs? Okay, let me see what are the inputs are given by the users. The participants I'm giving here array 3D from one colon from one. Yes, you're right. You're exactly right. You're on the right track. I'm happy to see you all send you any doubts. Let me know. I'm sharing the code in the chat Uh-huh. Are we good to proceed to discuss the introduction to pandas? Can you please confirm me? It's a new library. So pandas fundamentals of pandas we'll be discussing purpose of pandas features of pandas data structures like you have two different data structures which are very important as of now in the introduction part we'll be discussing only series part but we also have a very important data structure called as data frame which will model the data in the form of rows and columns. That's really very beautiful. We you'll get to know over a period of time. But uh now fundamentals of pandas. Pandas is an open-source library built on top of and is used for data manipulation. It introduces the data structures like data frames and series that make working with structured data more efficient. We have different purpose of pandas, features of pandas and data structures. Let's look into the introduction to series. when it comes to the series. Okay, sorry. When it comes to series, we are going to create and access panda series using the different methods. So, hope we are good to go ahead. and accessing. Creating an indexing panda series missing different. So as I told you pandas contains two data structures. What is that? Series and data frame. So series is a one-dimensional data frame is a two dimensional. Data frame contains the data in the form of rows and columns. And now I would like to create a series initially later we'll go to the data frame. Import pandas as ping where we'll be using the pandas. Then I'd like to create Bandas series I'm taking data equal to 1 2 3 4 5 and uh I would like to take the series is equal to PD dot series of data next I would like to convert create a panda series with a specified index. Folks please try to understand we have reported a pandas as pd. We have taken a data and I'm converting data to the series. Okay. Now a pandas creating a cond I'm taking index equal to A B C D And I I would like to also take with index is equal to P dot series of data comma index equal to Then creating a pandas series this folder. This is very important step. So I'm taking data dictionary creating a panda series from dictionary data dict equal column 1 column 1 2 D4 number Okay. and E col I would like to create this is a one dimensional okay from a dictionary So yeah series with index we are taking creating pandas series from a dictionary I have taken the data and what I do is equal dot is upper cases data comma index equal to this. So we are creating a panda says from a dictionary with the help of the data we have later I would like to create a series from a dictionary let's say index equal to d 20 I have five four values I take out of which I'm trying to get we have created a panda series from the and I'm trying to create a series series of So we have a two dimensional array is a data series series from dictionary and finally me. So let's uh forget this one. Creating panda series. So creating panda series from diction. So I'm taking index equal to A Panda says from a dictionary index is a mention. So whenever you are giving index and I'm saying select with index is equal to ed dot amount index equal to index later I'm creating Okay. Pas So if you want to create a if you want to store the series fun series or a creating a series you can create it from a dictionary data diction A col 1, col 2, col 3, and finally D col 4 col. Select from dictionary of data. So I would like to print and I would like to also print print with index. let us run and visualize invalid syntax colon. what you said? Like uh I see 1 minute we'll end the session for today. Meanwhile, I'm trying to remove this error series from dictionary. Hey, index right amount is not defined. So, this is our last demo. So if any error is coming shall rectify amount is not defined it's saying select with index filter series of index index no problem we shall clarify it later that's really challenging if you can clarify if you can resolve the error that would be really it you work on towards the resolving the error I shall also work on it meanwhile I would like to know how was the today's session folks how was the today's session uh I'm converting the document to the PDF can you please confirm me how was the today's session Actually today we were going at a very very snail pace slightly I would recommend you to support me in improving the speed. Hope you understand. And I would like to download the file from the cloud the practice labs. Anyways, if you can fix it, you fix it. Otherwise, I shall fix it later. I'm downloading the file where you can share it onto the elements. Excuse me. Can you please respond? How was the session today? And I I would recommend you to please submit the feedback and uh respond to me in the chat window as submitted and I'm uploading the today's file So if you want to access a a specific data from the series, we have created a series, right? Print series of when you say two, what is the value available? Let's see second index. Can you see there? In the series what is the second index 0 1 and second index right three you'll be getting three series with index of I say two no I say index b because a b cde e we have given right index is this one series with At B index, what is the value available? Two. So you'll be getting two. Hope you're getting any doubts. I would like to discuss a very realistic I take months Jan, March, April, May. revenue I would like to say $1,000 $1,500 $1,700 1600 and 2,000 and the revenue series we have been discussing introduction to pandas we shall look into the we have seen all this we shall look into the basic information in panda series right basic information in panda series like tail head shape describe and unique and unique basic information I would like to get first n rows is equal to series dot end of three So what is that you are getting? First three rows you are getting right. Similarly last end rows last n rows what does it give you? Series dot tail is a function which will take the n number and returns you the required number of close from the last 10. You can see that. Do you want even this uh one two lines code also? Later I would like to get the dimension. So dimension is nothing but row comma Dimension is equal to series dot shape D. What is this again? It has five rows is the meaning. It has five rows is the meaning. So to generate the descriptive statistics, generate descriptive statistics. stats is equal to series dot describe. What does that give you? It gives you count. It gives you mean, average value, standard deviation, minimum, maximum and inter quarter ranges. So that's what descriptive statistics it is giving you with the help of describe function unique values series dot unique Can you see all values only returns unique values? I want to find out number of inequalities. In a series you might be having lot of values and out of that I would like to find out number of inequalities. Series dot n unique facing five values here. Hope you're getting what we're doing. For example, you want head is right. It is giving top three values. What does it mean? If you want to view top three transactions performed recently, recent three transactions. Yes. Can make use of a head. Likewise, survey data you want to analyze. Yes, you can analyze and you want to know how many fields are there like what are the rows, what are the columns you have, you want to analyze, then you can go with shape. If you want to get the total analytics over the results data, you have exam results data and you want to get average score, highest score and how scores are spread out. Yes, you can use describe and unique. if you want to get the unique values or let's say in a system there are different roles available like system admin user guest so I want to view the different roles available and you can see with the help of unique And there are lot of records out of which I would like to find out how many different city people are there in my in the data. There are some certain employees available. So let's say thousand employees are available. I would like to find out to which different cities all these employees are available. So you can get unique any unique number of unique cities. So this way you can work with the different functions of later operations and transformations in pandas. let's see operations and transformations in one pass. So what are the operations we have? Operations and transformations in panda as a series are crucial for modifying, enhancing and cleaning data efficiently. They provide flexible flexibility to adapt to the data to specific analysis or visualizations preparing it for meaningful insights and ensuring data quality. So if you want to perform some kind of analysis over the data or pro some visualizations and prepare some meaningful insights then it is useful. So we shall see element wise addition. Element wise result series is equal to series plus series with the index series right then you can see what is that available it is showing you na Why? Because when you are adding series, you when you print series separately, yes, you will be able to print. When you are printing series with index, you'll be able to print. But these two doesn't have any common indexes. Hence, addition is not possible and you are getting not a number. If I want to apply, If I want to apply a function to each element, play function to each element. Let's say I have a series of values. I would like to square each value. So square series series dot apply lambda x colon x ^ of 2 and uh I'd like to print squared series. So you getting squared series earlier was 1 2 3 4 5. Now squared series we have applied a lambda function over each and every element. You are multiplying it x² you're doing right x into x and you are getting 1 into 1 1 2 into 4 likewise you're getting all squared values. Instead of one value and one is getting displayed instead of two value TW2 likewise three and four five I didn't mention so it is giving you n so this way we can map with the help of map function by mentioning a dictionary instead of one I'm asking asking it to display O1 instead of two I'm asking it to display TW2 likewise three this way we can create a mapping mapped series with the help of a dictionary I was expecting some outputs how you I would like to sort the series by Sort the series by values. Sorted series is equal to sort values. Generally if the series is not in a sorted order it will sort and distribute. Actually we have taken a sorted series since no effect is seen. That's it. Hope everyone is following. If you want these are one two lines demos and the very simpler ones and I hope you are comfortable following and how to check the missing values. check for missing values. So missing values is equal to series dot is null missing anything none of them is missing values. If anything is missing value you'll get a true here. No, everywhere you have values there's no missing values here if you want to handle if you want to handle missing values what you can do I would like to fill missing values with Okay. Specified value fill series is equal to series dot fill null value with the zero is the meaning then I can say fill it series. So anyways we don't have any missing values so it will not fill with any zeros but this is the actual process to fill the missing values and this is the actual process to get the missing Okay I shall wait here 1 to 2 minutes to make you comfortable before moving to the next section. So if you want to visualize these series transformations in a table format like Excel, we can convert them to the data frame and view them side by side. You can confirm me. Yeah, Sanjab your indexes should match. Then you can add Sanju, did I clarify about war? Are you comfortable? Okay, great. So, moving ahead. Next is all about querying. Querying a series. Querying is all about selecting and filtering data based on a specific condition in an essential aspect of is an essential aspect of querying a pandas scripts. And uh you can see some demonstrations here for quering operations. We shall take this panda series. I'm taking this panda series. this is a panda series. If you want you can check out the series, right? This is the series. What I would like to do, I would like to see I would like to get the elements which are only greater than 30. So select greater than 30. What do I do? Simple. I'm taking a variable selected greater than 30. Series of series greater than 30 can you see you'll be getting only over the series I'm quering saying that series greater than 30 what's happening it is only selecting the series which greater than 30 like 40 and 50 and you can see you are getting only 40 and 50 as a output. So I would like to give you the series actual series for which we are querying now. Okay, all of you complete it and post me the output. I shall wait here 1 minute so that we'll query few more things. Meanwhile, I have one query to be done by you. Selected equal to 20. So how do you do this? By referring to the above one, you try to do this. and only based on the your…

Transcript truncated. Watch the full video for the complete content.