Compaile

any more updates to this? running it on gpu would be very cool

First of all, format the code in a proper way. makes it so hard to follow. Also please don't create such bloated examples if you want a discussion provide a...

Again the algo is not wrong. Having a connection between 2 nodes does not mean they have to be in the same cluster. However, you can modify it a bit...

There is also an issue(cuda not dlib) that if you just upscale instead of doing a fixed size as target (I know letterbox and stuff) you need to relocate cuda...