Hacker News new | ask | show | jobs
by b112 1174 days ago
They could have trained, then modified code, repeat, to better enhance training in the current version.

Then after, train on raw data.