I want to know how to obtain the results that are presented in the paper. When I run the config file from GITHUB for ghz_346, the graph that is published, and what I get when I run are entirely different. I can understand that its because of the random weight initialization and also the order in which edges are removed. How to obtain the same graphs as you have got in the paper?
Originally posted by @Rithishbarath in #25
Originally posted by @Rithishbarath in #25