Fig. 6
From: Geometry aware physics informed neural network surrogate for solving Navier–Stokes equation (GAPINN)
two dimensional embedding of shape encoded latent vector k by means of t-SNE technique for visualization. Gray shapes indicating shapes from validation set and black shapes indicating shapes from training set. Under each shape is the mean squared error (mse) with respect to CFD-Simulations plotted (squares: mse u, triangle: mse v and circles: mse p). The color is linearly scaled between 0 and 1, the corresponding maximal values are msemax u = 0.15, msemax v = 0.07 and msemax p = 0.27