Search code examples
juliapcaeigenvector

Julia: pcaeig(X) yields "UndefVarError: fliplr not defined"

I am trying to get the eigenvector to do pca (principal component analysis). The package, DimensionalityReduction.jl offers a command that should do that very thing, pcaeig(X) where X is some matrix. My code is as follows

using DataFrames

using DimensionalityReduction

data = readtable("Midterm Data.csv")

T=size(data)[1]

n=size(data)[2]

erates = convert(Array,data[1:T,2:n])

eigvec = pcaeig(erates)

I do apologize that the formatting is bad, I don't quite remember how to put the code in a quote. Anyways, when I try to run this code, I get the following error: "UndefVarError: fliplr not defined". Now, to my knowledge, fliplr is a command used to flip a matrix (not a variable). It is also saying that the error is happening in the code for the package (not my code). Does this mean that I am out of luck and cannot use this package until it gets patched? If so, does anyone else know another method to get the eigenvector for pca?

Solution

  • As it says in the README of DimensionalityReduction, the package is deprecated:

    The DimensionalityReduction package is deprecated. It is superseded by a new package MultivariateStats.

    The package does not work on recent versions of Julia and will not be updated to do so in the future. Use MultivariateStats instead.