Figure 6 | Scientific Reports

Figure 6

From: Affine transformations accelerate the training of physics-informed neural networks of a one-dimensional consolidation problem

Figure 6

Comparison of AfPINNs and vanilla PINNs for TPM by mean (thick line), as well as minimum and maximum value as shaded area calculated over 500 runs each with intermediate values taken every 500 steps for 25,000 total epochs. MSE Loss, \({\mathscr {L}}_2\) relative error, and maximum of the absolute error for displacement u and pressure p are plotted with logarithmic y-axis.

Back to article page