imgcomp-cvpr icon indicating copy to clipboard operation
imgcomp-cvpr copied to clipboard

about soft and hard quantization

Open Linli621 opened this issue 4 years ago • 2 comments

hello, I don't figure out that when testing the model , why to use the hardout for the input of the P network .In my opinions ,the hardout and the symbols are equal in a sense .

Linli621 avatar Apr 09 '21 09:04 Linli621

I think one contains values at the symbols. e.g. if your quantization grid is:

levels = [0.2, 0.5, 0.7]

and your symbols are

sym = [0, 0, 1, 0, 2]

then we would ahve

hardout = [0.2, 0.2, 0.5, 0.2, 0.7]

The important bit being that we can have non-uniform quantization.

fab-jul avatar Sep 06 '21 18:09 fab-jul

Thanks a lot, I have figured it out.^_^

Linli621 avatar Sep 07 '21 11:09 Linli621