In several places where we're calculating internal/external diversity with incoming fingerprints, we randomly sample certain fingerprints, but ignore the very last one because we do a np.random.randint(0, n-1) instead of np.random.randint(0, n). This is probably a leftover from the R->python transition.
This changes the values slightly - only important for test cases here. No need to rerun any analyses of course.
In several places where we're calculating internal/external diversity with incoming fingerprints, we randomly sample certain fingerprints, but ignore the very last one because we do a
np.random.randint(0, n-1)instead ofnp.random.randint(0, n). This is probably a leftover from the R->python transition.This changes the values slightly - only important for test cases here. No need to rerun any analyses of course.