Overfitting, Underfitting And Data Leakage Explanation With Simple Example

No video

Overfitting, Underfitting And Data Leakage Explanation With Simple Example

Рет қаралды 23,181

Күн бұрын

In this video we will understand about Overfitting underfitting and Data Leakage with Simple Examples
⭐ Kite is a free AI-powered coding assistant that will help you code faster and smarter. The Kite plugin integrates with all the top editors and IDEs to give you smart completions and documentation while you’re typing. I've been using Kite for a few months and I love it! www.kite.com/g...
All Playlist In My channel
Interview Playlist: • Machine Learning Inter...
Complete DL Playlist: • Complete Road Map To P...
Julia Playlist: • Tutorial 1- Introducti...
Complete ML Playlist : • Complete Machine Learn...
Complete NLP Playlist: • Natural Language Proce...
Docker End To End Implementation: • Docker End to End Impl...
Live stream Playlist: • Pytorch
Machine Learning Pipelines: • Docker End to End Impl...
Pytorch Playlist: • Pytorch
Feature Engineering : • Feature Engineering
Live Projects : • Live Projects
Kaggle competition : • Kaggle Competitions
Mongodb with Python : • MongoDb with Python
MySQL With Python : • MYSQL Database With Py...
Deployment Architectures: • Deployment Architectur...
Amazon sagemaker : • Amazon SageMaker
Please donate if you want to support the channel through GPay UPID,
Gpay: krishnaik06@okicici
Telegram link: t.me/joinchat/...
Please join as a member in my channel to get additional benefits like materials in Data Science, live streaming for Members and many more
/ @krishnaik06
Please do subscribe my other channel too
/ @krishnaikhindi
Connect with me here:
Twitter: / krishnaik06
Facebook: / krishnaik06
instagram: / krishnaik06

Пікірлер: 56

@karthicktj1857 3 жыл бұрын

One of the best example forever I seen over fitting,under fitting,Data leakage.

@madhurajadhav3150 3 жыл бұрын

The analogy you used for explaining data leakage is so good that i will never forget what it is. Superb!!! i recently came across your channel and cant wait to watch all your videos. Thank you :)

@Krishna-pn5je 3 жыл бұрын

Sir, Thanks for the very good example . I just want to reiterate the above example with small changes. Valid Model -> A student whose performs well in his internal exams as well as his final exam IIT because he understood the concepts very well so he can imagine if the question asked indirectly. Overfitted model -> A student who performs very well in internal exams but he fails to perform best in the final exam IIT because he prepared internal exams based on previous questions that were asked in the internal test and did not understood the concepts very well. Underfitted Model -> A person does not perform well in the internal exams so he cannot perform well as in Final exam like IIT. Data Leakage :-> A person trying to attempt a exam for which he already knows question before the exam. and He cannot crack the other competitive exams .

@asadsaleem118 3 жыл бұрын

Never gonna forget this now. 👍🏻

@vaibhavyaramwar 3 жыл бұрын

Great Example and Explanation is really Simplified

@equiwave80 3 жыл бұрын

Hi Krish, To summarize we can say the below w.r.t Student scenario: Scenario#1: Ideal Student or Role Model Student with Conceptual learning hence ends up as a Topper!!! --> IDEAL MODEL Scenario#2: Bookworm or Mugging up expert who just memorizes the syllabus without really understanding the concepts --> OVERFITTING Scenario#3: Lazy Student who does not put any effort in preparing for the Exams hence ends up getting less scores --> UNDERFITTING Scenario#4: Paper Leakage resulting in temporary high scores but the student might end of performing badly in final exams --> DATA LEAKAGE Thanks Krish now I will never forget the Overfitting and Underfitting concepts!!! Regards, Deepak.

@nandhakishore8950 3 жыл бұрын

One of the best explanation for Data Leakage sir...

@hari141v 3 жыл бұрын

A nice and simple example for overfitting and underfitting

@shikhar_anand 3 жыл бұрын

Thank you Krish for taking up this example and explaining it.. Today I was asked this same question in the interview and I went blank and was unable to explain the concept properly.. This cleared everything and probably I won't forget this ever.

@yashiyengar7366 3 жыл бұрын

Beautiful analogy Krish, thanks for sharing it with us :)

@shahnawazkhan1636 3 жыл бұрын

Awesome I understand 100% please make videos on real life scenario like this video. Again nicely explained krish sir

@jifryissadeen5272 3 жыл бұрын

Thanks Krish.. Keep doing the good job.

@shubhamporiya1470 3 жыл бұрын

Hi krish Great explanation and a novel example for the terms to explain. 😇

@SnehaKumari-mb4px 2 жыл бұрын

Wow!!🤩 krish. So nice example

@pothusandeepkomal1006 2 жыл бұрын

wow that example was well placed krish ...good one

@ppsheth91 3 жыл бұрын

Very Nice Example Sir for overfitting, underfitting and data leakage, thanks..!

@MrAmarSindol 2 жыл бұрын

Very good example Krish, we will remember this forever, lol

@anaghadamame196 3 жыл бұрын

Amezing explanation 👌 Sir can you suggest some projects ideas which follows industry standard and add more weightage to our resume

@equbalmustafa 3 жыл бұрын

what if Accuracy on test data better than accuracy on train data???...plz let us know U explained very well...Thanks for making such wonderful content for us

@nabiltech1366 3 жыл бұрын

Yeah.Thats mean your model is perform well.But i dunno if u have data leakage

@shivambatra70 3 жыл бұрын

Awesome Explanation!!!

@pratikjadhav5244 3 жыл бұрын

what an amazing example

@zulfiquarshaikh3461 3 жыл бұрын

Remarkable explanation sir

@shirinsultana2809 3 жыл бұрын

execellent lecture of overfitting

@Neha-rx6zc 2 жыл бұрын

It's really awesome real world example sir. 👍

@karthicktj1857 3 жыл бұрын

Sir please make a vedio time and space complexity, of all algorithm ,most interviewer ask a question on working ,performing time

@TheFearmonger 3 жыл бұрын

if you go to te MIT open courseware site there are algorithm courses taught there for free, and you can even get a MIT certificate from it, and same for the harvard cs courses. Also if you have the ability to afford it, algoexpert is expensive but ive only heard good things. I hope these can help you as well, have a good day :)

@lucky_u4727 3 жыл бұрын

Thank you sir for your valuable class.

@ruvu9 2 жыл бұрын

Tq so much superb sir

@deepaknimbalkar8295 2 жыл бұрын

Very Good example given Sir Thank you guru.

@prakashsaravanan6613 2 жыл бұрын

Huge Thanks sir

@maulana6969 3 жыл бұрын

Sir, How do we recognize the data leakage problem... A good model and data leakage model have similar performance meteics...then how will we identify the data leakage problem for a model.

@nabiltech1366 3 жыл бұрын

I also wanna know

@mohammedameen3249 3 жыл бұрын

when you bring a new data to test the model you can check the performance metrics , if it is good ,then it is a good model otherwise not good and you have data leakage .

@adipurnomo5683 3 жыл бұрын

Data leakage happened if you put testing data on training data then your accuracy Will be increased because of that k fold cross validation is used to take this problem.

@ruvu9 2 жыл бұрын

Good example sir

@farhadkhan3893 2 жыл бұрын

awesome example ,

@universe3406 3 жыл бұрын

OP Krish !!

@nayanjain3594 3 жыл бұрын

Hi Krish, Thanks for the amazing explanation. I have one doubt... whenever I tried to get the accuracy of training data, I am getting as 1.00....why is it so?

@DeepakPandey-gs4gf 3 жыл бұрын

which algorithm did you use?

@nayanjain3594 3 жыл бұрын

@Faraaz Sheriff Actually first I have separated the dependent and independent variable then I have created train and test datasets...then I have applied random forest classifier model on training data(X_train, y_ train) ....After that I have tried to calculate the accuracy score of training dataset... whenever I am doing this I am getting the score around 1.0

@nayanjain3594 3 жыл бұрын

@@DeepakPandey-gs4gf Random forest, decision tree, linear regression

@SatyendraJaiswalsattu 3 жыл бұрын

Very well explained.

@jaiminshah143 3 жыл бұрын

Will you be adding this video in the Complete ML playlist or any other playlist ?

@sowmyatushar7487 3 жыл бұрын

very well explained !!!!!!!!

@tarung7088 3 жыл бұрын

Lets say tommorow is your semester exam and you havent studied anything in past. Now in one night you managed to mugged up everything in your head on considering that same questions will come tommorow. Overfit: the student who mugged up all the questions and answers word by word without understanding their gist. If a similar question appeared in the exam in a twisted way he won't be able to answer the question Underfit: the student who prepared very less for the exam and writes the same answer for every question well-fit: the student who prepares well enough to undrestand the gist of the question and answers. If a similar question appeared in the exam in a twisted way he will be able to answer the question.

@jayhingrajiya3703 3 жыл бұрын

Hello Krish. Congratulations! I heard you are getting married

@jameshunt1822 3 жыл бұрын

Disappointed. I thought you will show an actual world example. All you did was an analogy

@AmitSingh-wi8iv 3 жыл бұрын

Hello sir

@preetdahiya3012 3 жыл бұрын

Sir how can get data science job as fresher?

@preetdahiya3012 3 жыл бұрын

Or internship?

@adipurnomo5683 3 жыл бұрын

9:20 I thought students were get high score because random fluctuate statistic 😂. So we use statistic comparison classifier to test significant of the student