How to find Feature Importance in your model

[@justtohost3431]

What a relief to see this fantastic video on the question I've been asking myself for months!

[@ahmedamanibrahim192]

Nice tutorial! thanks for posting this

[@duykhanhang2940]

so touching for an excellent video

[@infinitygeospatial1972]

How about an writing an iterative function to do for a large dataset with large number of columns

[@MuhammadTahir-pb2xc]

Hey @Jonathan Perry.. Thanks for this nice video. I am subscribing your channel. cheers.. Keep posting awesome content

[@mzhichong2788]

Your content is so touching

[@jonathanperry8587]

Thank you! I do hope to have more up in the future. Time has been of short supply the last year.

[@user-ol7bq4ly1l]

Any chance you would do a video specifically for data pre processing on a csv file with a handful of features? Thank you!

[@ayieerosdi3629]

Mesej yang jelas, struktur yang jelas, mudah difahami, terima kasih

[@sotirispap906]

what happens if we have more columns ?do we need to repeat the same process several times?

[@jianjiang8234]

Great video! It's very helpful. Many thanks. I've noticed that sklearn has a 'Permutation feature importance' function. Will do the trick?

[@methembethomastshuma9587]

thanks for the information sir quite a helpful walk through

[@caassimbah2485]

really helpful video thank you.

[@wfpnknw32]

very useful, but are you able to upload a higher quality version i find it hard to read all the slightly blurred text for a long time. Great content though.

[@ichweinicht1858]

it is already available in 720p if u can't read in that u need a better pair of eyes

[@eatb67]

Hi Jonathan, Thank for your video! I was wondering, do you private tutoring in ML?

[@jonathanperry8587]

Hello Eliot, I would love to do this, but at the moment I do not have the capacity. I do expect summer time to open up more availability for me, but at the moment I would have to say no. Please stay in touch, and if you have any ML topics that trouble you, please post them, and I will add them to an on-going video list!

[@owusubright1046]

please leave the link here

[@wussboi]

You didn't even consider normalising each feature?

[@jonathanperry8587]

Thank you very much for your response, and of course you are more than welcome to normalize any feature. This was aimed to show the basic function calls and steps, not necessarily the proper theory or reasoning. Admittedly, I am an avid supporter of normalizing, but did not consider it in the video. I will work to put another video up discussing normalization and its own importance. Thank you again for the response, and I'll link the future video when up.

[@wussboi]

@@jonathanperry8587 no worries mate. Please continue creating great content for sharing. 🙌🏻

[@jonathanperry8587]

@@wussboi Thank you for the kind words, I will try to keep the videos coming and ever improving. If you have any suggestions, I am always happy to do what I can.

[@umakanta7]

Btw if you use any tree based model, you basically do not need to normalise any feature

[@danieldude15]

could you please share a link to this notebook?

[@danieldude15]

great video by the way :) Thanks

[@khuyenvu9625]

the sound is not good I am quite disappointed

[@andreamaiellaro6581]

i am getting curious error to me: when i define train_df = df.iloc[train] i get error: TypeError: unhashable type: 'numpy.ndarray'......

[@anhhoang-qu2ow]

the sound is not good I am quite disappointed