Y
Hacker News
new
|
ask
|
show
|
jobs
by
mentalgear
265 days ago
seems like they use the normal transformer architecture versus deep fold's more specialised machine-learning approaches.