Feature Elimination and Variable Importance in R with "caret" (2021)

[@mariabryant9240]

Hey Richard, Great video - sorry to see you stopped this series.

[@benxneo]

Hey Richard! Great video! Happy to see videos that cover R. I wanted to ask whats your view on tidymodels vs caret. Do you think its worth learning caret still? Or should a beginner learn tidymodels straight away? Hope to hear from you soon! Cheers.

[@RichardOnData]

At this point I recommend learning both (tidymodels I think is easier and quicker to learn), but caret is still more mature at this point. My answer might change in a year or so. Got a video on this coming soon!

[@miyinzi]

Hey Richard, Can you do a similar video when doing RF Regression. This one is classification. If you have a link, kindly share.

[@RichardOnData]

That's a good idea to go deeper into the difference between those two. Great video topic that I'll try to do in the near future!

[@poojamahesh8594]

i need to find the accuracy of the model after finding the variable importance...please tel, me how to do

[@sofiasalgueiro3454]

Thank you so much for the video! I have done a classification model that has 4 outputs, so when I try to extract the variable importance with varImp I can generate that plot, but I can't see the importance for each output. So for the most important variable I can't understand if it's importance because it helped to predict output 1 or 2 or 3 or 4, or if it is important because it helped to distinguish between what 2 or 3 outputs. I'm new at this, so I wanted to know if there is any way I can have this information. Thank you! Kind regards

[@alessandrorosati969]

I try random forest method but i have accurancy equal zero

[@mrgermannudelman]

oh my god.😲