                            
                            StarNet++ v2.0.2
							    Finally!
                            
-> As before: Only TIF files with 16bit per channel are supported. <-

              -> Yes, AVX is still needed. Sorry! <-
   -> If the previous version did not work - this won't either. <-
   
          -> TO BE USED FOR ASTRO IMAGE PROCESSING ONLY <-

                            -> Have fun! <-
							  

This will be announced on cloudynights.com, so you might want to
ask for help there if you face any problems (I suspect that GUI
might be problematic for some people, but can't do anything about
it in advance)!

----------------------------------------------------------------------

Hopefully, the GUI is pretty self-explanatory. A couple of important
points:

1. The output is not guaranteed to have the same profile as the
   original, so the colours might appear different. If this happens,
   all you need is to just open the image in the software of choice
   and restore the profile.
   
2. This version will sometime leave patterns around small stars.
   It is often prominent with low noise images. If you face this,
   you might want to upsample your image (2x for example) and these
   artefacts will go away.

----------------------------------------------------------------------

The code is quite heavy and will require decent amount of memory
(with about 8.2 MP images I tested it with its up to 3 Gb at the
start, about 1.1 Gb during the most runtime), so make sure you are
not running a potato. Obviously, I am not responsible for any loss
or damage if your computer will freeze, blow up, or collapse the
Earth into a black hole as a result of running this software.

This brings us to the following remarks:

First, the supplied tensorflow.dll was compiled for 64bit systems
and won't work in 32bit Windows, which means if you have Windows 
XP installed, you will most likely not get it running.

Second, the tensorflow.dll was compiled to support AVX instructions
and will only work if your CPU supports these AVX instructions.
Which means that CPUs made before 2011 will not work.

----------------------------------------------------------------------

Dependencies. The code depends on C API of Tensorflow and libtiff.

    http://www.libtiff.org/
    https://www.tensorflow.org/install/lang_c

Make sure that tensorflow.dll, tiff.dll and starnet_weights.pb
are in the same directory as starnet++.exe or somewhere in $PATH

Huge thanks to Daniil Goncharov for his hello_tf_c_api project:

    https://github.com/Neargye/hello_tf_c_api
    
part of which was used in this project.

----------------------------------------------------------------

Please see LICENCE file for legal information.
By using this software you agree to the terms.