Detection accuracy is not equal to darknet

[surajitkundu29]

I have used my own weights and configuration file. But while I am testing with new images the darknet can detect 4 objects where keras can detect only 2. Do you know the reason behind this ?

[SagiShaashua]

I'm having the similar issue - I am using the same weights and the same cfg as the Darknet one... and i can't get the same results. For one example while Darknet detected a car with 99% the this one had only 26%... what is the difference between the two?

Did you find a solution?

Thanks