notabigfish

I have the same question. Is there any update?

I met the same problem. mAP of ZF is only0.0956, while mAP of VGG16 is 0.6922. Did anyone solve this?

@Coastline2018 Hi! I've seen that you use Ubuntu and it works. Have you figured out a way to compile them on windows? Thanks!

Sorry, I found the mistake in my code. For fully connected layer, `network_regularizer` should be `GroupLassoFlopsRegularizer`.

@eladeban Dear Elad, thanks for the detailed explanation. I appreciate this opportunity and would like to try to contribute. Honestly I really don’t know what the first move should be...

Hi, after reading the above explanations, i'm still confused, sorry for this... `cam_extrinsics[:3, :3] = np.matmul(np.matmul(np.matmul(np.linalg.inv(R_vg), cam_extrinsics[:3, :3]), R_vg), R_gc) ` In the formula above, there are 3 matmuls: 1....

> I have solved it. thanks anyway @zhang3689 Hi, how did you label the detection box and get the mAP result? Thank you!

Thank you!

@LossNAN it's ok, thanks! And which resnet in the original paper did you use? Resnet 152 or ?

@kamauz Hi, do you find the solution? I also changed the network structure and faced the same problem as yours.