Neural-IMage-Assessment icon indicating copy to clipboard operation
Neural-IMage-Assessment copied to clipboard

Published 20 hours ago • verified publisher icon yunxiaoshi

Learning Rate Setting

Open mr-Mojo opened this issue 5 years ago • 1 comments

Hi, I am experimenting with the NIMA implementation for a scientfic project! In your Readme, you say that "The learning rate setting differs from the original paper. I can't seem to get the model to converge with momentum SGD using an lr of 3e-7 for the conv base and 3e-6 for the dense block.". Which settings did you use? The defaults in argparser are set to the 3e-7, 3e-6, so that's why I was wondering!

Thank you :)

mr-Mojo avatar Nov 05 '19 10:11 mr-Mojo

0.005 for the dense block and 0.0005 for the conv base gives you a reasonable result. But I suspect it is not optimal, maybe do a grid/random search

yunxiaoshi avatar May 27 '20 06:05 yunxiaoshi