Timeline for When computing mAP for an object detection model, how many detections should one consider?
Current License: CC BY-SA 4.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 19, 2021 at 23:44 | vote | accept | IntegrateThis | ||
| Jan 19, 2021 at 23:44 | answer | added | IntegrateThis | timeline score: 0 | |
| Jan 12, 2021 at 5:33 | history | edited | IntegrateThis | CC BY-SA 4.0 |
added 1 character in body
|
| S Jan 12, 2021 at 0:05 | history | bounty ended | CommunityBot | ||
| S Jan 12, 2021 at 0:05 | history | notice removed | CommunityBot | ||
| Jan 4, 2021 at 3:17 | history | edited | IntegrateThis | CC BY-SA 4.0 |
added 515 characters in body
|
| Jan 4, 2021 at 3:15 | comment | added | IntegrateThis | @tchainzzz I've thought about this a bit more and added some ideas. | |
| Jan 4, 2021 at 3:07 | comment | added | chang_trenton | Have you looked into $mAP@k$? This is $mAP$ conditioned on the top-$k$ highest-confidence predictions. A suitable value for $k$ is going to be very task dependent; as long as you're transparent about which $k$ you choose though, that should be OK. | |
| Jan 4, 2021 at 0:00 | history | tweeted | twitter.com/StackStats/status/1345882684451672065 | ||
| S Jan 3, 2021 at 21:39 | history | bounty started | IntegrateThis | ||
| S Jan 3, 2021 at 21:39 | history | notice added | IntegrateThis | Draw attention | |
| Jan 1, 2021 at 23:35 | history | edited | IntegrateThis | CC BY-SA 4.0 |
added 122 characters in body
|
| Jan 1, 2021 at 23:26 | history | edited | IntegrateThis | CC BY-SA 4.0 |
deleted 14 characters in body
|
| Jan 1, 2021 at 21:38 | history | edited | IntegrateThis | CC BY-SA 4.0 |
deleted 3 characters in body
|
| Jan 1, 2021 at 21:28 | history | edited | IntegrateThis |
edited tags
|
|
| Jan 1, 2021 at 20:49 | history | edited | IntegrateThis | CC BY-SA 4.0 |
deleted 8 characters in body
|
| Jan 1, 2021 at 20:26 | history | asked | IntegrateThis | CC BY-SA 4.0 |