Hacker News new | ask | show | jobs
by qorrect 1520 days ago
He might mean features like dropout, or out of sample training, randomness that's introduced during training. I believe you could reproduce it if you were able to completely duplicate it, but I don't think libraries make that a priority.