Timeline for Consistently inconsistent cross-validation results that are wildly different from original model accuracy
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 3, 2017 at 17:34 | history | edited | pmp | CC BY-SA 3.0 |
added code
|
| Mar 3, 2017 at 17:11 | comment | added | pmp | This doesn't really explain why shuffling the rows would results in such completely different cross validation scores. | |
| Mar 3, 2017 at 17:07 | history | edited | pmp | CC BY-SA 3.0 |
Added more of my code
|
| Mar 3, 2017 at 16:59 | history | edited | pmp | CC BY-SA 3.0 |
Added more of my code
|
| Mar 3, 2017 at 0:13 | comment | added | epattaro | its a little difficult to evaluate what you can be doing wrong without a sample code. But if i had to guess i would say your first score/validation is being done with trainning data, hence the high score. | |
| Mar 2, 2017 at 23:55 | history | edited | pmp | CC BY-SA 3.0 |
improved formatting
|
| Mar 2, 2017 at 23:52 | review | First posts | |||
| Mar 3, 2017 at 1:39 | |||||
| Mar 2, 2017 at 23:43 | history | asked | pmp | CC BY-SA 3.0 |