Skip to content
This repository has been archived by the owner on Oct 11, 2023. It is now read-only.

AP when training with SVM #39

Open
meirpr opened this issue Mar 20, 2019 · 0 comments
Open

AP when training with SVM #39

meirpr opened this issue Mar 20, 2019 · 0 comments

Comments

@meirpr
Copy link

meirpr commented Mar 20, 2019

I have completed all the steps of Part1 using 2000 ROIs. However my AP's for the classes are much lower than written . They should be:
image
but I get:
Number of rois before non-maxima surpression: 1583
Number of rois after non-maxima surpression: 137
Evaluating detections
AP for orange = 0.1364
AP for eggBox = 0.0341
AP for joghurt = 0.1364
AP for ketchup = 0.2727
AP for squash = 0.3455
AP for mushroom = 0.0000
AP for water = 0.0227
AP for mustard = 0.0000
Mean AP = 0.1185

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant