Ridge and Lasso regression. To prevent overfitting that can occur in linear regression. Ridge regression: L2 regularisation, you add lambda*(slope)^2, a penalising parameter Lassos regression: add lambda|slope| as penalising parameter to the residual error.aka L1 regualrisation This prevents overfitting as well as feature selection is done
Overfitting (Low Bias, High Variance): The model fits the training data well but fails to generalize. Underfitting (High Bias, Low Variance): The model is too simplistic and doesn't fit the training data well or generalize.
Model may not be able to perform well on new datasets. Example: Assume a Kid having a math test who over practiced 100 Questions at home. But in test, he has been asked some new kind of Questions. How likely he will be able to perform well? Very low right. Thats why.
Ridge and Lasso regression. To prevent overfitting that can occur in linear regression.
Ridge regression: L2 regularisation, you add lambda*(slope)^2, a penalising parameter
Lassos regression: add lambda|slope| as penalising parameter to the residual error.aka L1 regualrisation
This prevents overfitting as well as feature selection is done
Hi Kirsh, this hindi channel is gold for data science aspirants, I request please make more videos on this hindi channel.
I think For underfitting -
High bias and low variance.
Please check it once.
Thanks.
Yes.
yes
@@Devavrat2005 no you are wrong
Yup
bro this is just mind blowing explanation this is the first time i have understood something so clearly
Best Video sir Thank you for the giving precious knowledge for free👍👍👍👍
Thankyou sir for the beautiful content..your videos are very informative and helpful. 😊
Hi krish. Thanks for all your amazing sessions. Please add continue with more SQL videos
Amazing content sir waiting for the next video.
Thanks. You are doing a fantastic job. stay blessed, and greetings from Oman.
Thanks sir for providing such a good content 😊
Thankyou for the great explanation!!!
amazing hindi session it was very easy to understand ... plz explain linear regression,ridge and lasso python code in hindi
High level of teaching
Sir can u explain more on how L1 used for feature selectin bez in both their is vector summation and thus squaring the coefficients
that was freaking great
And what would be the formulae for updation of the parameters?
How do you calculate lambda constant value ?
Great video! What role does the Lambda value play here? I mean, what effect does it have on bias and variance if I change (increase or decrease) it?
Thanks for machine learning
Thank you Sir
sir please upload the RandomForest algorithms
Bhaiya ji x theta n y i. R they specific to machine learning because econometrics me hum simply ise beta hat likh dete h
this video is a recommendation for bias vs variance and not Ridge and Lasso BAM!! DOUBLE BAM!!
BAM @statquest
but what that lambda means?
he directly put value but if someone asked what is lambda then how we have to answer? pls guide
that lambda is regularization parameter.
read more about regularization you will get to know
How do we select the value of lambda?
Using cross validation
Overfitting (Low Bias, High Variance): The model fits the training data well but fails to generalize.
Underfitting (High Bias, Low Variance): The model is too simplistic and doesn't fit the training data well or generalize.
Here is one correction:
1. Overfitting: Low bias and high variance.
2. Underfitting: High bias and low variance.
amazing sir
sir plz provide notes and code doc
👌
Secondly, why overfit is a problem?
Model may not be able to perform well on new datasets.
Example: Assume a Kid having a math test who over practiced 100 Questions at home. But in test, he has been asked some new kind of Questions. How likely he will be able to perform well? Very low right. Thats why.
Can anyone help: How to check accuracy for train data nd Test data so that we can know overfitting and underfitting condition ?
if you know answer for this question then please let me know.
Find accuracy score for (y_train, y_pred) for training data.
And accuracy score (y_test, y_pred) for testing data.
Bro tum bhut ganda padha rhe ho, sach mein, conceptual knowledge bilkul nhi h tumhe.
Yes, I think you are right.
Kisi bhi cheez ki defination ni likhwaya hai.
Direct graph and. Formula likhwa diya hai.
kuch bhi smz ni ayya ..............kya yrr.....pta nai kya padya
Thank you sirjee
Can anyone help: How to check accuracy for train data nd Test data so that we can know overfitting and underfitting condition ?
Find accuracy score for (y_train, y_pred) for training data.
And accuracy score (y_test, y_pred) for testing data.