mescla.emma.endmembers.suggest_endmembers¶
- mescla.emma.endmembers.suggest_endmembers(U_samples, U_candidates, n_endmembers, labels=None, max_combinations=20000)[source]¶
Rank every combination of candidates by how well it bounds the samples.
Exhaustive over combinations, which is affordable for the handful of candidates a real study has. Ranks first by the fraction of samples enclosed, then by the volume of the simplex – among sets that enclose everything, the tightest is the least presumptuous.
This suggests; it does not decide. An end-member has to be a water that exists and that your conceptual model can name.