Definition Of Bias And Variance In Machine Learning- Interview Question

Subscribe To ‪@krishnaikhindi‬ for videos in Hindi For Python and Data Science
-----------------------------------------------------------------------------------------------------------------------
Recording Gears That I Use
shorturl.at/wzI68
---------------------------------------------------------------------------------------------------------------------------------------------------------------
The Oneneuron Lifetime subscription has been extended.
In Oneneuron platform you will be able to get 100+ courses(Monthly atleast 20 courses will be added based on your demand)
Features of the course
1. You can raise any course demand.(Fulfilled within 45-60 days)
2. You can access innovation lab from ineuron.
3. You can use our incubation based on your ideas
4. Live session coming soon(Mostly till Feb)
Use Coupon code KRISH10 for addition 10% discount.
And Many More.....
Enroll Now
OneNeuron Link: one-neuron.ine...
Direct call to our Team incase of any queries
8788503778
6260726925
9538303385
866003424

[@sidharth1452]

Hello guys,
Please don't get confused.
For Training data :-
Good accuracy --> low bias
Bad accuracy --> high bias
For Testing data:-
Good accuracy --> low variance
Bad accuracy --> high variance

[@thepresistence5935]

Thanks for this 💃🏽

[@rafibasha1840]

3:14 it should be low bias when model is performing well

[@VARISHROCKS]

I too have the same doubt !

[@palashjyotiborah9888]

Me too. Doubt

[@saitarun6562]

yess training data(high accuracy)=low bias(low error rate) training data(less accuracy)=high bias(high error rate) here you can call bias as error rate.

[@satyanarayanavempali8511]

Me too that doubt

[@mohamedhafidi4911]

you goddamn right

[@vishaldas6346]

One can only understand these conpects better while implementing the same into their respective work. Your videos are worth watching and I urge people to practice such concepts into their respective projects to get clear idea.

[@saitarun6562]

thanks kirsh for this video model 1(example for over fitting) and model 2(example for under fitting) finally model 3 is perfect model to use.

[@sumanreddy4165]

Sir, could you make videos on algebra and calculas, and how to code these in python to use these effectively in machine learning.

[@karrmanbhatia4261]

Krish i think you mentioned wrongly in case of bias at 3.20min.... when model is performing well on training data it means error is low and this is low bias case but you said when model is performing well on training data its high bias
high bias means high training error

[@suriyaprakashkk9365]

3:14 low bias ......not high bias.....bias and varience is just a error respect to training dataset and test (Validation) data set.....when model perform well, it means low error.

[@saitarun6562]

yes training data(high accuracy)=low bias(low error rate) training data(less accuracy)=high bias(high error rate) here you can call bias as error rate.

[@suriyaprakashkk9365]

@@saitarun6562 yeah .....yes that is exactly what I'm saying ......and training time error is known bias and testing time error known varience.....I hope this is correct.

[@ikram_6024]

Thank you for this clear presentation. I have a question about low variance variable, how to find low variance variable (threshold)
, why and when we should remove this variables ?

[@venkatesh532]

Finally i broke blackbox of Bias & Variance.. Thank you Krish 🙏🙏

[@AmanKumarSharma-de7ft]

In model 1 the accuracy training is high still u r saying low bias .....whereas u said high train accuracy means high bias

[@vishaldas6346]

When traing accuracy is high, bias is low. Bias is the error term at training level and is inversly proportional to the accuracy.

[@AmanKumarSharma-de7ft]

@3:10 it's mentioned that high accuracy means high bias ...is that a mistake?

[@shivshankarshukla1509]

@@AmanKumarSharma-de7ft I think so !!

[@mukul4816]

@@AmanKumarSharma-de7ft that was by mistake

[@AmanKumarSharma-de7ft]

Ohh okay thanks a lot ...it was confusing because he never makes mistakes😅

[@ParasProgramming123]

Sir agar m hp Pavillion aero 13 laptop leta hu to ye acha rhe ga long term k liy kyu k models to Google colab m train ho jay gy ap ka experience ka khehta h?

[@anandvyavahare2031]

3:14 I think you said it opposite.

[@JaiRudraNath]

I think you explaining in white board the traditional way is far far better than digital board..

[@prudhvimunjeti9232]

Once again rocked 🔥🔥

[@rafibasha1840]

Low bias/low variance :When model is performing well on both train and test
High bias :When model is neither performing good at training/test
High variance :Performing well on train but not on test

[@deen4803]

Sir please continue mySQL series also

[@ManojKumar-ge4dj]

To be corrected....definition of high and low bias in training data set

[@narayanbabubharali9846]

Nice