Lower mAP when training on VOC

[Kevinz-code]

Hi, thank you for providing your source code of the paper,
Actually, I run you training code of VOC dataset for 3 times, but always get the Total mAP at around 77.7mAP. I notice in your paper "Table 1", the mAP of your "DB-focal" is 78.94. I was wondering if I did something wrong and why this is the case ?

Below is the log-out information of your training code.
Split: head mAP:0.7344 acc:0.8429 micro:0.6096 macro:0.5392
Split:middle mAP:0.8359 acc:0.9375 micro:0.6400 macro:0.6527
Split: tail mAP:0.7651 acc:0.9400 micro:0.5492 macro:0.5305
Split: Total mAP:0.7771 acc:0.9101 micro:0.6024 macro:0.5698

Thanks in advance.

[wutong16]

Hi @Kevin655 !

Thanks for your attention to our work.
The issue is probably due to the sensitivity to the randomness because of the small data scale.
Please try another random seed, e.g. --seed 1, which produces a total mAP of around 79.0 here on my server.
And we would also consider improving the robustness of our method to randomness.

[Kevinz-code]

Got it！ Thank you for providing the reproducing method!