Why does the pca function in MatLab and on mathworks indicate different outputs?

I am trying to run a simple pca on a data set using as indicated below as part of a function

[~, scores, ~]= pca(X)

However, I keep getting the following error

Error using pca (line 151)
Input NCOMP is missing. Type: 'help pca'

Error in function (line 22)
[~, scores, ~]=pca(X);

It won't format the outputs how I want them. Does this have something to do with the fact I have PLStoolbox?

Answer 1

[Steven Lord]

You have a different pca.m that is taking precedence over the pca function included in Statistics and Machine Learning Toolbox. In release R2019b line 151 is a comment listing one of the references for the function. To confirm this look at the output of this command:

which -all pca

[Adam Rish]

How do I switch to the preferred/deisred function?

[Steven Lord]

I would ask the author of the other function(s) why they're shadowing the function from Statistics and Machine Learning Toolbox and probably suggest they move their function into a class or package folder.