Handling Missing Values in Pandas Dataframe | GeeksforGeeks

[@GeeksforGeeksVideos]

WISH TO CONTRIBUTE VIDEOS ON GEEKSFORGEEKS? Please submit this Google Form - docs.google.com/forms/d/e/1FAIpQLScSVFlE-stz69M_l46xcwZlVRzDxT_91dlc-px5ytr7QHYmkQ/viewform

[@hmuhamadzulfitri]

Addition:
1- In case, you want to **impute** the null values of numerical column, you can check the distribution of the values first; If it's not skewed, you may use mean. In case it is skewed, you may use median.
2- Generally, before we **delete** null values, we can check the proportion of the null values. If it's less than 5 % of the total dataset, we can opt to delete the null values.
However, always take the context of our data before making any decision with the null values. 😬

[@KeerthivasanSundararaman]

got a solution after about 30 mins of search.. Thank you made my day..

[@mrtbhdr]

geeksforgeeks never disappoints, thanks for this!

[@darksector1389]

You, sir, are amazing and a savior to my project 🙏

[@-DanishShafi]

Thanks brother.
The video was amazing, it cleared all my doubts.
Rise and shine🎉

[@chiemelieonu8846]

Thank you. The class, for me, was very insightful and rewarding. Worth my time.

[@senglay5729]

Clear explanation and easy to understand. Thank you!

[@lukmanmanggo]

Hi. very easy tutorial bro. As very beginner on python I can get it easy. Salute

[@Gouthami-xr4eh]

Thank you .very nice and clear explanation .Easy for everyone to understand.thank you sir .

[@raumitraj8628]

Bhut acha conceptual content hai yrr. It's really awesome!!!

[@programsolve3053]

Very nice and helpful lesson. Thanks for sharing. 🎉🎉🎉

[@sougatamitra5112]

One of the best explanations!! kudos

[@CuriousLearner_05]

Thanks for the clarification , helped in clear a problem I was stuck in

[@noorelahi4792]

Really amazing sir.
after that wanna enroll for DS course.

[@vndsanju9865]

chala chala TQ anna nuvu chepindhi chala thoraga ardham chesukunanu and chala bhaga ardham iendhi

[@MrKbforex]

Great content..Exactly what I have been looking for... Thanks

[@mariamartin6205]

Thanks so much, this was really well explained and helpful.

[@daysirivas30]

Woooo!! Súper easy explained!!

[@erison6791]

Thank you it is wonderful explanation. You answered my question perfectly.

[@priyankajoshi1953]

Very nice explanation 👍

[@as8401]

very very clear thank you so much for this fantastic video...

[@saiprateek6643]

Great resource for EDA ☺☺

[@bosunomo]

Great content, thank you so much.

[@ketanverma7839]

Informative, can you please also make a video telling how to handle null values using ML algorithms ?

[@poornimachougale8903]

thanks for the informative video..🙏

[@kushangshah-be7hn]

It was help ful, thanks bro

[@jagadeshkumar6174]

Thank you, very clear and nice tutorial......

[@the-ghost-in-the-machine1108]

clear explanation. Thanks

[@motivation_for_everything470]

really nice
session

[@nyashagracenhandara7757]

thank you the explanation is clear and helpful

[@hibbajameel8779]

Thanks! You help me in my project ❤

[@Ayanshedipelly2312]

We have to take care of data types also before filling with any value

[@VoidmanEric]

so at 21:50 limit direction = 'forward" does work similar to method = 'pad"?

[@user-eg3sg1bz7g]

this is so good and easy to understand thank you

[@yohanneskebede1573]

well explained. Can you make a video to fill in Na values by generating random numbers around the mean?

[@abebebirhane7332]

good presentation and nice knowledge

[@abdoulkarim1468]

thank you ; very good explanation

[@falcan7752]

Nice one Mate!!!!

[@radhakrishna.kalluri7192]

Good job... keep up good work

[@AteetNP]

when should we drop a row?, i'm always fill the values with mean, median, mode etc, because rest of the columns seems importand.

[@abouteverything1573]

Clear explain ❤️

[@moatazsakr5435]

thanks so much on this video ❤

[@mikefranko2832]

But does it make any sense, to fill NA values with the next or the previous value? Or even more useless to fill it with a 0 or 5? Why would you do this?

[@johnbainbridge1931]

I need intelligent filling, eg filling NaN with the mean of a data observation's category. E.g. car weight is missing, but we know it's SUV, so we apply average SUV weight.

[@Chris-si4ox]

this is brilliant thanks

[@khushaldodeja8267]

Clean. Thanks!

[@raghunathanp3734]

Great content

[@tarblood]

thnx it is a good explanation

[@v44s82]

#2
f = lambda df=df_rec: df.isna().sum(axis=0)
result = f(df_rec)
print(result)

[@Uzi5272]

thank you for this video

[@srikanthkoltur6911]

It was truly great
thanks

[@bennguyen1313]

How would you fill empty values (NaN or NA) in a list or dataframe with the last known good value?

[@MelissaLu-v5t]

Hi Sir, great video, one question, when fill in the mean(), or max() for certain column, the entire whole dataframe was filled in with that mean() even the categorical column, can I just fill in the mean for that column but keep the other columns unchanged?

[@mohdzohairrizvi4377]

did you get the answer to your question?
I also needed the same question to be answered.

[@jodysidney9761]

@@mohdzohairrizvi4377 Create a dictionary eg fillna({'colunmname1': 0, 'columnname2':5}). It will replace nulls in column1 with 0 and nulls in column 2 with 5

[@milliekim5072]

Thank you so much!

[@navaneethangn1215]

Thanks Dude!

[@bagushari1886]

How to fillna using interpolate method in the some column in once time?

[@v44s82]

#1
df_rec = pd.read_csv("recommendation.csv")
print(df_rec)
print(df_rec.dtypes)

[@csit3093]

Super class

[@nikomangrustini7553]

I'm sorry Mr, can you make tutorial handling missing values with interpolasi linier method?

[@kakashiaravind221]

Sir, Let us the above as same example, now i need to save the dataset(after handiling the missing value) as csv file

[@jimymunandar5319]

how about fill the null value if the datasets is categkrical

[@drewbal2265]

for i in range(108):
print("Thank you")

[@haripriya9101]

can we have a mean value for pad or bill

[@djaysse]

Thank you, your explanations are very clear! How can I follow more courses by Akshit Madan?

[@salahotman8450]

Thanks

[@ajaykushwaha-je6mw]

Hi Team,
I want to replace missing value in original df. I used below code but it is not working, kindly help.
df[(df['Age'].isnull()) & (df['Pclass']==2)].fillna(29)

[@uniqueuday8569]

Filling and replacing is same i think?

[@hemkumarreddymaryada8140]

Find the number of rows with missing values means
Can I use df.isnull().sum().sum() in Python ?
Can anyone , please answer me ?

[@animehunter2169]

what if the missing values are defined by "-"

[@sravankumar3411]

Bro hw remove the $ symbol in column

[@rajulgupta3614]

you should explain in which scenario which type of filling method is appropriate or suitable this is not a good video
for example if we have outliers then we can not use mean

[@v44s82]

A LEUZ
L fichier .zip eli baathhouli ghazi mafihech questionet, fih les fichiers csv wl fichier jupyter mais maghir questions, EKTEBLI LES QUESTIONS 1) ... 2).. F KOL COMMENTAIRE WAHDOU WALA MABADDHOM WALA KIMA THEB wena taw nekteblek b ay façon
juste hajti bel question chnia 1) .. 2) .. 3) ..

[@sudeep2580]

print"(Thanks for showing so many methods")

[@venkateshpeddinti5601]

👏

[@hadirdammak5055]

hey

[@v44s82]

ebaathouli les questions
ektebli juste 1) .. wala 2).. commentaire w to njeweb west l comment
l fichier eli baathou ghazi mafihech questions ferghin 1) 2) 3) ...

[@fernandojackson7207]

I used the nan at the Indian restaurant and the waiter brought some bread.

[@RitwikKapur]

luv u

[@v44s82]

#3
df_rec = df_rec.dropna(subset=["recommendation"])
df_rec = df_rec.dropna(subset=df_rec.columns)
df_rec = df_rec.drop_duplicates()
print(f(df_rec))
df_rec

[@mohdzohairrizvi4377]

how to replace the value of only a particular column by the mean of that column?