Excessive runtime using tune.block.splsda()

That is odd. They’re all relatively small datasets so shouldn’t be taking too long. If you run the same code on your laptop and the cluster, and it only works on the former, it may be an issue with your Linux cluster. If the progress bar is not showing at all (and you’re sure you’ve set progressBar = TRUE), then this is likely an indicator that your issue is machine specific.

Unfortunately, if you’re unable to reproduce the issue with your second machine then I am very unlikely to do the same. Hence, I’m not sure what I can do for you.

What I want is a biological signature of the components that best represent each group, i.e: Males Obese and Nonobese, Females Obese and Nonobese, PCOS females obese, and PCOS females non-obese.

The plotLoadings() function will aid in visualising this. It can colour the bars by which class is associated with the maximum/minimum mean/median (set by the contrib and method parameters respectively). Look here for more info

The greeds of the tuning component I am not so sure how can I put in them. On the docs, it says that sequences of like seq(1,50,5) should work, but on other docs, it says like (10:50,seq(51,60,2)) so I am not so sure what I can do.

I shall quote myself below from this post which I’d recommend have a read through. Tuning keepX in an iterative manner - by decreasing the range and increasing the granularity of the grid at each step - will save time.

Ensure when you tune these models that you use an adequately high number of repeats (ie. nrepeat = 100 ). As this may take a lot of time, I’d also suggest doing the tuning over multiple steps - increasing the resolution and decreasing the range of the grid at each iteration. Eg. Start with test.keepX = seq(10, 150, 10) and based on the output (lets say it selects 50), then undergo tuning again with test.keepX = seq(30, 70, 5) ; rinse and repeat.

If you have any more questions about analysis and usage of the package I can help. However, I believe your issue of excessive runtime on your Linux cluster is something that I cannot aid in diagnosing. Best of luck.