Figure 4
Results of the NN 2-128-32-8-1 neural network (see Fig. 3). This NN was trained and tested using a data set generated from \(n=3\) different \(\phi _p\) and for each \(\phi _p\) \(N=300\) different \(\dot{\gamma }\). Top-left: loss as a function of number of epochs; Top-right: predicted versus actual values of the training (blue) and test-holdout (orange) set; Bottom-left: predicted versus ground truth of the test set containing unseen \(\phi _p\); Bottom-right: examples of ground truth (actual) shear rate dependent viscosity (\(\circ\) (seen \(\phi _p\)) and \(\square\) (unseen \(\phi _p\))) and predicted shear rate dependent viscosity (\(\bullet\)). The colors indicate different \(\phi _p\).
