Some questions on the code.

[sfleisch314]

I've been going through your training notebook and have a few questions. BTW: Thank you very much for this project it is helping me to understand the model.

• Do you implement the "Bag of Freebies" augmentations?
• Also, the CIoU, GIoU losses?
• Similarly: Where is the NMS implemented? I see NMS_THRESHOLD assigned in cell 12 of YoloV4_tf but not ever checked. Is NMS implemented as part of the custom loss code?

Best regards.

[RobotEdh]

Sorry for my late answer.

NMS is implemeted in the function do_nms with currenlty a threshold = 0.5.

I didn't tried to add Bag of Freebies" augmentations for the training, but with tensorflox you have several possibility that you can define during the training.

For CIoU, GIoU losses, i didn't have time to study how to implement them but happy to get some input on this topic.

Eric

[sfleisch314]

Thank you for the response. I just wanted confirmation that my understanding of the code was more or less correct.

Best regards,

Steve Fleischman From: RobotEdh [email protected] Sent: Thursday, September 24, 2020 2:51 PM To: RobotEdh/Yolov-4 [email protected] Cc: Stephen Fleischman [email protected]; Author [email protected] Subject: Re: [RobotEdh/Yolov-4] Some questions on the code. (#5)

Sorry for my late answer.

NMS is implemeted in the function do_nms with currenlty a threshold = 0.5.

I didn't tried to add Bag of Freebies" augmentations for the training, but with tensorflox you have several possibility that you can define during the training.

For CIoU, GIoU losses, i didn't have time to study how to implement them but happy to get some input on this topic.

Eric

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/RobotEdh/Yolov-4/issues/5#issuecomment-698553893, or unsubscribehttps://github.com/notifications/unsubscribe-auth/APMEBPOYLHTQV5ISF2UE2N3SHOPJVANCNFSM4QJ2WGWA.