CornerNet-Lite icon indicating copy to clipboard operation
CornerNet-Lite copied to clipboard

Published 20 hours ago • verified publisher icon princeton-vl

How long to train the CornerNet_Squeeze?

Open ybai62868 opened this issue 5 years ago • 7 comments

Hi, author! I wanna ask some questions about the training speed of your CornerNet_Squeeze with the default hyper-parameters in your configuration. How long did you take to get the results (on the coco dataset) in the paper compared with yolo_v3?

ybai62868 avatar Apr 27 '19 05:04 ybai62868

We trained CornerNet-Squeeze from scratch and the training took around 2 weeks to finish.

heilaw avatar Apr 28 '19 21:04 heilaw

We trained CornerNet-Squeeze from scratch and the training took around 2 weeks to finish.

hi,How much is suitable for loss?

lililiiiiiiiiii avatar Apr 30 '19 00:04 lililiiiiiiiiii

Why the validation loss is much larger than the training loss?

gtwell avatar May 22 '19 01:05 gtwell

How long to train the CornerNet_Saccade?

SeeeeShiwei avatar May 25 '19 10:05 SeeeeShiwei

@heilaw I want to finetune the CornerNet_Squeeze from coco to pascal voc, and i just want train the classifier, which index or name of layers should be train?

tc-qaq avatar May 27 '19 07:05 tc-qaq

@heilaw I want to finetune the CornerNet_Squeeze from coco to pascal voc, and i just want train the classifier, which index or name of layers should be train?

hg

gtwell avatar May 28 '19 09:05 gtwell

We trained CornerNet-Squeeze from scratch and the training took around 2 weeks to finish.

If I didn't wish trained CornerNet-Saccade from scratch, what kind of network model should I choose as the basic model?

czhi3 avatar Dec 19 '19 07:12 czhi3