The Original Mean and Standard Deviation for Tabular?

Hi all, is there a quick way to gather what the mean and standard deviation were for your data? I’m trying to recreate some columns to go further analyze the results.

Thanks!

For anyone wanting to find them, they’re pretty buried. I had to do the following to grab our training datasets:

procs = learn.data.train_ds.x.processor[0].procs[2]
procs.means
procs.stds

Is there a faster way to natively to grab them? If not mabye a get_stats() function?

1 Like