Closed rzshrote closed 1 year ago
I believe this is caused by the DenseBreedingValueMatrix class when it centers and scales breeding values. If the scale is zero (caused by fixation of all alleles), this causes a division by zero error and induces a NaN value.
Not sure about the origin of this bug. Was using
algmod.gebv(pgmat).descale().sum(1).mean()
and a NaN was introduced. This bug cannot be easily replicated.