#25 Missing Values Handling in Pandas | Data Science for Beginners in Tamil

Key Takeaways

Welcome you all to this episode 25 of this data science with machine learning course data cleaning operations filtering sorting selections group by functions Monday to Friday 6:30 p.m. Saturday inspirational stories data science one Make sure you're available on that particular time in case what is we will definitely see whether we can take you in and provide a bridge and give you some job opportunities and work or not. We will be giving you a call. for comry to put it in the comment section. Let them also come and study library input and output. Data frame indexing, selection, sorting, filtering. Data cleaning operations. Handling missing data. Different types of questions in the complete episode number real time. foreign. So, don't worry. deter cleaning operations different data structures twodimensional data frame filtering selection and sorting questions Find out five new questions along with script answers. That's really great. Chapter five. This is going to be the last chapter in pandas. Chapter five. Um we'll just go with missing values. How to deal with missing values. How to deal with missing values? employee data cvate open and close it's a function employee okay import the data Excel index number always starts from Huh? Okay. So, By default, not available. Not a number. Not a number or not a value. Okay. Let's say data scientist maybe Average question. So imagine problemat. So this is very serious. First business understanding data understanding our operations. department. Okay. Okay. Fine. Step one. Okay. Step one. Step one. I'll call to my client. I'll call to my client and check with him if he can give the right data. Write data. Right. First step. Step two. Fine. Great. Super. In case in case client is asking us to perform the data cleaning operations. Seven observations. Imagine it. Same entry. Imag first step. So step one data clean suggested always suggested always recommended. Step one remove duplicate entries or we can say records. Records. First employ bridge employee data bridge employee data prompt [Music] duplicates duplicates Stop duplicates open and close return with duplicate rows removed. Okay, that's great. Duplicate. Super bridge employment by default False. And the values datable. It will impact on the complete operations. The complete operations whatever it's performed will impact the original the original data set. step by step. Okay. Duplicate step find out find out the null entries. Four. Find out null entries. Employee data. Copy. Total number of number of Okay. Duplicate model based on age salary prediction. Okay. Imag [Music] Okay. Okay. The age. Okay. So first step in the particular dataf Because machine learning model particular entry For example, ProTit bridge employing values. Okay. for the first placeal by default. False. Bridge. Record it. Step three. Record. in a document. Document record in a document. Okay. Client gave totally seven observations. observations out of which out of which three entries three entries are removed? Three entries are removed because those entries those records do not have relevant do not have complete information. model. I'll send this. Okay. Send this to client and get the approval to proceed for model building or model training. Step four. Step four, get approval. Okay. Ctrl C the dot is null dot sum we are good to go first approach best approach duplicates records duplicate records data handling. technicular column. Average column, Fore you won't get a good impression on me. So first data cleaning process. So that will be the right approach data scientist because I'm having super confident everybody will be working but in the particular practice episode mostly of real time teaching real like this is what the starting point next level of data clean this is your a John let's make a simple notified. See you all in the next episode. Bye. Cheers. [Music] [Music]