Caffe-DeepBinaryCode icon indicating copy to clipboard operation
Caffe-DeepBinaryCode copied to clipboard

Question about Objective Function

Open RishabGargeya opened this issue 8 years ago • 2 comments

Hi,

Does this implementation scale the objective loss function? If you don't scale each of the three loss parts, aren't they implicitly weighted then based on their raw values (since you set alpha beta and gamma to 1 in your experiments)?

Please let me know if I'm missing something. Thanks for your time!

RishabGargeya avatar Jul 13 '16 09:07 RishabGargeya

@RishabGargeya: Current github implementation does not scale the objective function. We have conducted the parameter selection and study the influence of individual terms in the learning objective, but our latest manuscript is still under review.... We will update our codes and paper after paper accepted!

kevinlin311tw avatar Jul 14 '16 04:07 kevinlin311tw

HI, I have some question on the image rotations hash projection loss function,the loss calculated process is mismatch with the formula(8) in the paper, i want to know how to understand the relationship between this computing method and formula(8). Best wishes for you look forward for your reply.

YIFanH avatar Feb 10 '17 02:02 YIFanH