Machine learning feature engineering: Label encoding Vs One-Hot encoding (using Scikit-learn)

[@TheAnbohan]

Hi, thx for the tutorial, however the sklearn doc specifies that you should use LabelEncoder for target not for features: "This transformer should be used to encode target values, i.e. y, and not the input X."

[@MrMadmaggot]

What is target values? the labels (y's)?

[@havryliuk]

You could use integer encoding for the dependent variable but you won't be able to use one-hot encoding because you would be predicting more than one feature, not sure if that's possible.
Well, encoding is really used for independent features (x) as well because machine learning models can't be trained on labels (strings) but only on numbers.

[@updatascience]

You are right, thanks for your comment, in my new video I explained this: I have uploaded a new video for you to cover this: When to apply each categorical encoders? Practical examples ua-cam.com/video/JbaELidWCQw/v-deo.html

[@havryliuk]

I suppose it would have been useful to mention when to use each type of encoding and what are their pros and cons

[@cristianofroes4681]

I was expecting the same, to know when use each one and why.

[@updatascience]

I have uploaded a new video for you to cover this: When to apply each categorical encoders? Practical examples ua-cam.com/video/JbaELidWCQw/v-deo.html

[@updatascience]

@@cristianofroes4681 I have uploaded a new video for you to cover this: When to apply each categorical encoders? Practical examples ua-cam.com/video/JbaELidWCQw/v-deo.html

[@stand4justice4867]

Thank you very much! the video was clear, direct and informative :)

[@todorp4056]

Thanks, nice tutorials

[@sasidharansathiyamoorthy6918]

Can someone explain on what basis the labels are generated? Like why did Staten Island get 3, Manhattan 2 etc?

[@updatascience]

Hey Sasidharan, it is generated by the fit_transform function, and the function is assigning them a unique represented number; this is the whole idea of label encoding, which is mapping each value into a unique number

[@musyo005]

In this technique, each label is assigned a unique integer based on alphabetical ordering. Bronx(0) comes before brooklyn(1),manhattan(2),Queens(3), Statens Island(4).

[@_danfiz]

@@musyo005 thank you!

[@RamonX69]

@@musyo005 how to specify each label for each value?

[@cristianofroes4681]

I was expecting to learn when and why use each one also the main differences. Anyway, thanks for the great class.

[@updatascience]

I have uploaded a new video for you to cover this: When to apply each categorical encoders? Practical examples ua-cam.com/video/JbaELidWCQw/v-deo.html

[@ameerabdul5444]

thanks for tutorial ....i like the way u explained

[@updatascience]

Happy that you have liked it!

[@yahyamlaouhi9508]

nice video keep going

[@updatascience]

Thank you!

[@pkstock372]

airbnb.shape( ) , it would be more easy to understand .

[@updatascience]

Thanks for commenting

[@saunitmarolia3901]

Nicely explained

[@irfanfulari2122]

Focus on indian students, you have good scope...