Timeline for What does Logits in machine learning mean?
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 1 at 14:11 | comment | added | Rune | ...and a link to that question from here, please? | |
| Mar 11 at 8:11 | comment | added | noe | Hi @user1195883, could you please create a new question on the site with your doubt? | |
| Mar 11 at 7:56 | comment | added | user1195883 | This got me thinking: why do we actually need to normalize at all? Since both training and prediction/interference seem to work fine without it, why and when do we have to normalize? | |
| S Apr 10, 2023 at 8:59 | history | edited | Lynn | CC BY-SA 4.0 |
In the original answer the sum in the denominator of the function definiton of the softmax goes from j=1 to K. The author of the answer always uses N for the number of classes in all other instances. The upper limit for the sum is whatever variable is used for the total number of classes+edit speech
|
| S Apr 10, 2023 at 8:59 | history | suggested | kltobias | CC BY-SA 4.0 |
In the original answer the sum in the denominator of the function definiton of the softmax goes from j=1 to K. The author of the answer always uses N for the number of classes in all other instances. The upper limit for the sum is whatever variable is used for the total number of classes+edit speech
|
| Apr 9, 2023 at 18:55 | review | Suggested edits | |||
| S Apr 10, 2023 at 8:59 | |||||
| Aug 4, 2022 at 9:08 | history | edited | noe | CC BY-SA 4.0 |
added 99 characters in body
|
| Jul 28, 2022 at 9:03 | history | edited | noe | CC BY-SA 4.0 |
added 77 characters in body
|
| Jul 28, 2022 at 8:57 | history | answered | noe | CC BY-SA 4.0 |