Random seed
qdread opened this issue · comments
For now, there is no warning given or display of random seed so results of null models will be slightly different each time the functions are run, unless the user sets the seed beforehand. Also, the algorithm for hypervolume is stochastic so even the point estimate will differ each time unless user sets a seed. I think it might be a good idea to print the seed being used and/or prompt the user to set one, so that we enforce the results being consistent.
I now fixed this so that you can set a random seed within Ostats() and Ostats_multivariate(). If you don't, it will warn you and set a seed based on the current time. The commit is e16f4b4