Hello @amdegroot,
thank you for making your code available. I am currently working in the ssd_keras port. However, we are missing the MAP score and I saw that you have already calculated yours. Do you think you could pin-point which code did you use to evaluate your SSD port.
Also I saw that you are missing the data_augmentation part. Maybe you could take a look here. It is a python generator, it is currently missing the crop transformation but it has helped me reach a better loss.
Thank you!
Has this MAP score calculation portion been updated ?
Hi @kimdon20,
No I still have not figured out (also because I stopped searching for it) where to calculate the MAP score, but it could still be very helpful :) (@amdegroot)
+1. Hope for some mAP calculation code.
Hey guys, just posted the mAP calculation code in eval.py. I have also posted a link to some pytorch weights that achieve a slightly higher mAP than the original caffe weights now that data augmentation and the fully training from scratch pipeline is complete. Sorry this took so long.
Most helpful comment
Hey guys, just posted the mAP calculation code in eval.py. I have also posted a link to some pytorch weights that achieve a slightly higher mAP than the original caffe weights now that data augmentation and the fully training from scratch pipeline is complete. Sorry this took so long.