Handling missing data | Numerical Data | Simple Imputer

Simple Imputer is a practical solution for filling missing numerical values in a dataset. This method replaces missing entries with the mean, median, or a specified constant, providing a straightforward approach to address and mitigate the impact of missing numerical data in your dataset.

============================
Do you want to learn from me?
============================

📱 Grow with us:

⌚Time Stamps⌚

00:00 - Intro
00:37 - Handling Missing Numerical Data
03:33 - Mean / Median Imputation
07:55 - Code Demo
17:15 - Imputation using SKlearn
20:15 - Arbitarry Value Imputation
22:40 - Code Demo
25:57 - End of Distribution Imputation
30:09 - Outro

[PS_nestvlogs]

Nitish sir ek aap dekhna ek aisa time aayega jab Data science bolne par aapka channel CampusX hi hum jaise bande dekhenge.. so rich content.. apka videos dekh ke feel hua that agar pata hota pehle main course enrol nahi sakta sirf aapka channel se Data science seekh jata. you are an inspiration to everybody from this domain. kabhi channel band mat karna sir

[katadermaro]

Man, I can't stop commenting about how thorough your content is. Every other video on youtube covers this topic in 5 min videos. Thanks a lot for taking the time to talk about all the nuances.

[rachit_]

From now onwards, I will comment on every video I watch of yours.

[ajaykushwaha-jemw]

Sir, app ki jitni bhi tareef ki jaye wo kam hai. Aap bahot bahot accha padhate hain.

[ayesha]

I'm literally SO happy I found this video, I was trying so hard to find any resources that explained when to impute with what, and you explained it SO well here. Thank you so much

[namanjoshi]

can't believe the depth and how easily i understood everything ... this content is top tier

[AltafAnsari-tfnl]

This channel contains the best content with notebook for each topic and in-depth intuition. Keep it up.

[JACKSPARROW-chjl]

Thanks a lot bhai, god will fulfill all your wishes 🙌🙌🙌

[devgupta]

At 10:57, sir has mistakenly filled missing values of Fare with the meidan_age and mean_age. It should be median_fare and mean_fare, respectively.
The correct code is there in Github repo

[learningpoint]

i watched you SVM kernal trick video and after that i moved toward channel for looking brand new content to check your improvement in content, i feel that you improved too much, best wishes for you

[charansai]

Nithish bhayya I’m commenting on every video, but still you deserve the applause for detailing, im looking for taking the enrollment in the 2.0

[MeetShingala-unzl]

I wanted to ask if your are also supposed to fill the null values with median also in the X_test?!

[shreyasur]

Sir cross validation par ek video banaiye na pls

[MonkeyDLuffy]

I first like the video, then watch it.... I am pretty sure the content is awesome😁

[namanjoshi]

I have a doubt, if we have large field of numerical data missing like 10% < then what to do, which method to go for i mean. Btw great video sir !

[ali]

If the data missing is less than 5% we use mean/median in numerical.
If the data missing is more than 5%? then what do we do?

[jimmysandhu]

Hi,
As u always say ke fit the imputer/encoder on test data and transform both test and split. But isn’t there a chance of Data leakage with that. As explained in the course of kaggale data lekage happens if we do this thing .

[gabriel]

How do ya'll access English subtitles for this?

[mohitkushwaha]

Why cant we fill the missing data before split???

It would have become easier for us, else we have to repeat the same process for test data also.

[sanjanauprety]

sir the github link is not working.. its disabled at the moment..