'How to deal MAP(Mean Average Precision) if there are multiple predictions on same object
In my results, Cause i skip NMS process,There are multiple predictions of different confidence score on same object.
If all predictions exceed IOU threshold, Does MAP Metric count TP(True Positive) several times on same object?
Now for my evaluation i use pycocoapi tools
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|