Machine Learning Tutorial Python - 17: L1 and L2 Regularization | Lasso, Ridge Regression

Thanks so simple ❤😊

That guy doesn't know what regularization is. That thing u've called overfittng maybe solved with degree reduction and this is not overfitting. For linear regression u may see overfitting in data with a big number of fuatures and relatively small number of samples. Exactly here we need to use l1/l2/elastic regularization but and in this case we may apply PCA and problem will solved without l1/l2/lasso.
Learn ml, noob)

Nice explanation .. Adding to that
L2 Ridge : Goal is to prevent multicollinearity and control magnitude of the coefficients
where highly corelated features can be removed by shirking the coefficients towards to zero not exactly zero , stability and generalization.
L1 Lasso : Goal is to prevent sparsity in the model by shirking the coefficients exactly to zero , importance in feature selection, preventing overfitting..

The best of two worlds wow!

Hello Sir
why did you noy fill the distance parameter with mean value?

Clean, crisp and crystal clear, I was struggling to understand this from a long time, your 20 mins video cleared it in one attempt, thanks a lot💌💌

When I am creating dummies, it is showing that the Suburb column is of type NoneType() and no dummies are getting created. What can be the problem?

Thank you for this video. Very straightforward and comprehensive ❤

sir can you provide ppt and jupyter notebook link of above used resources?

I can understand it now, thanks to you 🥳

is there any algorithm using which we can determine the unimportant features in our datasets?

I really love learning from your Videos, they are pretty awesome.

Just a concern, as in Line 11 we ran a missing value sum code where the Price Stated, 7610 and in the next line that is Line 12, we have dropped the 7610 rows, isn't it?

Also, what was the other option if we would not have dropped the valued, can we not divide the data set and treat 50 percent of the missing values in Price and as a train dataset by imputing mean, and run the test on the missing price values.

I am not sure, even if this is a valid question, but I am a bit curious.

Also, what was the scope for PCA here?

How do you already know that it will overfit,,, thag means we have to check on every algorithm that score compulsory,,, where i learn ml they never told this ,,,

I really love your content….. You change lives❤❤❤

Great video.
However, It would have been better if you had provided the justification for assigning Zeros to few NaN values and giving mean to frew records. I know "its safest to assume" butt hen I believe in real world projects we cannot just assume things.

To me filling the "BuildingArea" feature with the mean when around 2/3 of your data points have NaN sounds like a bad idea. On the contrary you dropped the "Regionname" while only three data points have NaN values. I'd drop the 3 corresponding data points instead and keep the feature "Regionname", which is one of the most important features in determining the price of a house.

Maybe in the Cost formula, the indices for summation should be different (in general): for the MSE term the sum should be over the entire training dataset (in this case n), and the sum for the regularization term should run over the number of features or columns in the dataset

ho to ccomputer gradient of L1 reg its not even differentiable

Hello,
can you put the link of github?
Thank you

Hi bro I tried all the possible ways to get the data but I cannot and even tutorial.

Thankyou very much sir

Sir, i can't find link Belbourne_housing csv .

Thank you. This is very helpful.

Nice video....good lesson......funny enough i see my house address in the dataset

So are l1 and l2 polynomial regression models?

First when you apply lasso, you apply it apart from the first linear regression model you made right?
Which means applying scikit Lasso is like making a linear regression but with regularization or it is applied to the linear regresion from the cell above??

So what if I use a knn or a forest?

I think one must not use those imputations(mean) before train test split as it leads to data leakage, correct me if I am wrong.

Excellent Tutorial, Thanks.

Can you please provide the jupyter notebook link for this piece of code sir?

My lasso regression is getting wrong results. It is giving all coefficients as zero except the constant and R2 score as --0.001825328970232576. Someone please help.

Hi...The equation, shouldn't it be : Theta0 + Theta1.x1 + Theta2.square (x1)+Theta3.cube (x1) rather than Theta0 + Theta1.x1 + Theta2.square (x2)+Theta3.cube (x3) because we have only one x feature ?

2) the Regularization expression (Lambda part), my understanding is that we should not take "i & n" , rather we should take "j & m" etc. The reason is that in first half of equation, we took "i & n" for number of rows whereas in second half, we need to take number of features, so different parameters should be used.

Please correct me if my understanding is wrong.

Amazing. But how to select best alpha value?

good theory!

Amazing sir thank you so much

Note for myself: This is the guy... his videos can clear doubts with codes.

cool

great video, thanks!

could you plz share the code and data set ?

achine learning concepts and practicals made easy, Thank you so much Sir

I really liked your way of explanation sir

Cool video

thanks sir

what is dual parameter and please explain what is primal formal & dual

Always excellent lessons, thank you

Such a great video!! I was struggling to understand regularization and now it's crystal clear to me!

I tried Linear Regression on the same dataset but it scored the same with Ridge and Lasso why?

Nice Explanation. Also Recommended to play on 2X

thank you ! this video save my exam :)