Y
Hacker News
new
|
ask
|
show
|
jobs
by
elandau25
1778 days ago
That's a good idea. It will likely be trickier to apply a method like this for text. Decomposition of the problem is less obvious than with vision tasks.