Excuse me, but can you share the dataset you used for training?
#2
by ckw1140 - opened
This comment has been hidden
Sorry but I deleted the images. I believe I found them on Wikipedia Commons
There are images in the source repo but I believe those are just examples
This comment has been hidden
Sorry but I don’t remember if there was any pre-processing beyond downsizing the images. I believe there was typically 100-300 per class.
ckw1140 changed discussion status to closed