search

weiliu89 / caffe

Caffe: a fast open framework for deep learning.
http://caffe.berkeleyvision.org/
Other
4.77k stars 1.67k forks source link

About the 'variance' of prior boxes #342

Open yw155 opened 7 years ago

yw155 commented 7 years ago

Hi, @weiliu89,

I see a parameter of 'variance' in the prior box layer. And in the line 27 of detection_output layer, there is a 'detection_output_param.variance_encoded_in_target()'. I have not seen this function in the file. What is the default value of this function? Do we really used this parameter in the last outputs?

weiliu89 commented 7 years ago

variance_encoded_in_target is a parameter, not a function.