This is a follow-up of #56. I verified that by passing the seed to sklearn.model_selection.train_test_split the test results are consistent even when np.random.seed is not initialized. I left its initialization just in case, with an updated comment.
I also tried to pass the rng directly, but got this error: ValueError: Generator(PCG64) at 0x7F1758766A40 cannot be used to seed a numpy.random.RandomState instance.
This is a follow-up of #56. I verified that by passing the seed to
sklearn.model_selection.train_test_splitthe test results are consistent even whennp.random.seedis not initialized. I left its initialization just in case, with an updated comment.I also tried to pass the rng directly, but got this error:
ValueError: Generator(PCG64) at 0x7F1758766A40 cannot be used to seed a numpy.random.RandomState instance.