Timeline for Is it correct to evaluate Neural Network after a fixed number of batch-updates, rather than at the end of epoch?
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 7, 2022 at 18:48 | comment | added | étale-cohomology | @generic_user like in bootstrapping | |
| Feb 20, 2018 at 14:19 | vote | accept | user2740947 | ||
| Feb 20, 2018 at 10:23 | answer | added | Jan Kukacka | timeline score: 2 | |
| Feb 19, 2018 at 18:00 | comment | added | generic_user | Sure, why not? There is nothing magical about epoching, and I'd not be surprised if it's not even optimal to organize batches by sampling without replacement. | |
| Feb 19, 2018 at 17:52 | history | edited | user2740947 | CC BY-SA 3.0 |
edited title
|
| Feb 18, 2018 at 19:39 | review | First posts | |||
| Feb 18, 2018 at 21:06 | |||||
| Feb 18, 2018 at 19:37 | history | asked | user2740947 | CC BY-SA 3.0 |