Feature Encoding 101: Prepare Data For Machine Learning

[@Neuraville]

Great breakdown of how to transform features into usable numbers for better model performance.

[@menshift]

I love your 101 series....very clearly bullet-pointed.

[@fgrillo123]

Thank you, I really enjoy your explainations😁

[@Raminber]

I think it would be great if you can show how to do this in a sklearn pipeline. As far as I understand, this is the proper way to do this.

[@ChezBing]

Frequency and target encoding should only be done using the training sets !

[@abhay9994]

Good point, Can you explain why? I have some thoughts 1) After deployment while making real predictions model will only get actual values 2) lf your testing set is true representation of training set won't the frequency and target encoding values be same for both training and testing. So does it really make difference ? 3) While testing should we pass the actual value of that column like salary and won't it will confuse the model and give us lesser testing score. 4) Can we do something like store these values from training and use them ?

[@sagartalagatti1594]

bro, great video... It'd be more helpful if you could include chapters using timestamps, so that we can skip right to the feature encoding we want to see/revise later...

[@mugomuiruri2313]

good watching from africa

[@abhay9994]

Thank you 🙏😊😀😄🙂🤠

[@RahulVyas-ds4uv]

I love you my bro 💗💗

[@عبداللهعمران-ع6و]

Hello can you make big projects such as robot work in a car factory

[@Carbon_Toaster]

MATLAB users catching strays 😂

[@Somens8Dworld]

no js 😢