thanks for the vid...question...if i wanna improve prediction...how to know which variables to take out and which ones keep? there must be some logic or indicator or smth not just a pure guess
Great video. How do you get the function scores at the end of the process? For example, if I had 5 groups I wanted to classify using this supervised classification technique. Typically it will generate one less function than the number of groups (n=4). I am a total R-studio newb and would like to know how to get these scores?
The video is very well explained till you start talking about the interesting part that is the Quadratic discriminant analysis. From there you rush a bit.
I will never understand why people opt to not use caret when it makes everything so much cleaner and easier for you. Everything you showed here could be done more easily and more elegantly by using the functions from the caret package, which also uses CV very conveniently
Omg thank you so much I have been stuck in the same LDA problem for an hour, you helped me to solve it!!
Absolutely amazing! Thank you so much.
Many Thanks!
you can use logistic regression for any number of levels
You're right ! :o
thanks for the vid...question...if i wanna improve prediction...how to know which variables to take out and which ones keep? there must be some logic or indicator or smth not just a pure guess
Great video. How do you get the function scores at the end of the process? For example, if I had 5 groups I wanted to classify using this supervised classification technique. Typically it will generate one less function than the number of groups (n=4). I am a total R-studio newb and would like to know how to get these scores?
Nice !!
Thanks for sharing! I have a question: How did you choose the set.seed number? does it depend on sample size?
It can be any number and it doesn't matter. The main point is to just reuse whatever seed you pick if you want to replicate your results!
@@mathetal :) thank you! your video is really helpfull
all good except plots dont work for me :(
is this bayer LDA
The video is very well explained till you start talking about the interesting part that is the Quadratic discriminant analysis. From there you rush a bit.
ga ngerti mbak
I will never understand why people opt to not use caret when it makes everything so much cleaner and easier for you.
Everything you showed here could be done more easily and more elegantly by using the functions from the caret package, which also uses CV very conveniently
suwun mbak, saya ga jadi ga ngerti
lu goblok sih