[1803.09820v2] A disciplined approach to neural network hyper-parameters: Part 1 -- learning rate, batch size, momentum, and weight decay

[1803.09820v2] A disciplined approach to neural network hyper-parameters: Part 1 -- learning rate, batch size, momentum, and weight decay ▾ https://arxiv.org/abs/1803.09820v2

Added

2022-05-27 13:56

First Archived

2022-05-27 15:56

Last Checked

2022-05-27 15:56

Tags (09820v2)

Status

archived

Saved

✅ 10

Errors

❌ 0

Size

1.5 MB

🗃 Snapshot: `[1653659773.141625] a25d06ad-4a41-4d0a-a401…`

./singlefile.html

Chrome > SingleFile

./output.pdf

Chrome > PDF

./screenshot.png

Chrome > Screenshot

./arxiv.org

Wget > HTML

🌐 web.archive.org/web/...

Archive.Org

🌐 arxiv.org

Original

./headers.json

Headers

./output.html

Chrome > HTML

./readability/content.html

Readability

./mercury/content.html

Mercury

./media/*.mp4

Media

./git/*.git