I have two sets of N-dimensional arrays that I need to somehow compare and get a value representing how similar they are. In statistical terms, this is a two-problem fitness problem when the hypothesis is that two arrays are derived from the same parent distribution.
To simplify, imagine two-dimensional arrays like the ones shown below:
for which I need to imagine a number that quantifies how similar they are.
Is there a package pythonthat provides such a statistical test? I am open to using numpy, scipy, scikit-learnetc.
Add
scipy, , , , , 1-D: scipy.stats.ks_2samp. R ks package, kde.test. , , python.