4 ms·
It’s quite unlikely that training data will include duplicate repositories or even forks, that alone would surpass the published dataset sizes.
by ricardobeat 11mo ago
It’s quite unlikely that training data will include duplicate repositories or even forks, that alone would surpass the published dataset sizes.