My original use of sample was in the training phase of the neural network when I had relatively perfect images to train with and did not want to nor know how to create imperfect images. So I chose to artificially create imperfect images. Perhaps I should have used sample1 but I elected to use sample2 (my sample). I have yet to test my neural net with real imperfect cases and may be in for a real disappointment regarding its accuracy then. I am hoping that the dramatic 1:16 compression will overcome the real imperfections.
On Wed, Jan 2, 2013 at 12:41 PM, Raul Miller <[email protected]> wrote: > Which leads back to why I was asking "why?": I do not have a deep > enough understanding of your goals to distinguish between sample1 and > sample2, and I do not understand the value of the x argument in your > original definition of 'sample'. > > -- > Raul -- (B=) ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
