Figure 5From: CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial NetworksThe Minkowski functionals of the convergence maps, which are sensitive to the non-Gaussian structures. We carried out the measurements on 100 batches of generated maps (6400 in total) and compare them to those of the validation maps. The functionals are evaluated at 19 thresholds and shown here are the bands of \(\mu \pm \sigma \) at each threshold. The dashed lines represent the mean μBack to article page