AdamW and Super-convergence is now the fastest way to train neural nets
244 by tim_sw | 26 comments on Hacker News.
Subscribe to:
Post Comments (Atom)
Featured post
Popular Posts
-
Snap’s Spectacles v2.0 are faster, pricier, and as pointless as the first By Abhimanyu Ghoshal Snap has launched its second-gen Spectacl...
-
Predicting Stock Performance with Natural Language Deep Learning 145 by prostoalex | 49 comments on Hacker News.
-
Put Down the Pitchforks; Amazon Didn't Steal Your Dunning-Krugerrands By Comments
No comments:
Post a Comment