Fig. 24From: Deep convolutional architectures for extrapolative forecasts in time-dependent flow problems2D River problem, \(L_2\) relative error of the auto-regressive predictions with increasing time for the CNN modelBack to article page