By what amount can we reduce dimensions using DCT?

by user5398   Last Updated October 10, 2019 03:19 AM

I have a numeric dataset of 1000x100(1000 vectors with 100 components) and I'm supposed to reduce its dimensions using PCA and DCT.
I was able to reduce from 100 dimensions to 83 dimensions via PCA with 95% variance.
But I'm not sure on what the appropriate amount of reduced dimensions would be with DCT. We can use Eigenvalues to calculate data loss in PCA, what is something similar in DCT?



Related Questions


Updated July 16, 2015 17:08 PM

Updated March 11, 2017 12:19 PM

Updated May 23, 2019 21:19 PM

Updated May 04, 2017 05:19 AM

Updated April 28, 2016 08:08 AM