Figure 17
In both graphs, the horizontal axis is labeled “Epoch” and has markings ranging from 0 to 400 in increments of 50 units. Graph 1: The graph is titled “Offices Accuracy Curve.” The vertical axis is labeled “Accuracy” and has markings ranging from 0.4 to 0.9 in increments of 0.1 units. The graph shows 2 curves. The first curve for “Train” starts from (0, 0.4), rises concave down in a zigzag fashion passing through coordinates (100, 0.76), (300, 0.85), and terminates at (394, 0.85). The second curve for “Valid” starts from (0, 0.4), rises concave down in a zigzag fashion passing through coordinates (37.6, 0.71), (200, 0.86), and terminates at (400, 0.86). Graph 2: The graph is titled “Offices Loss Curve.” The vertical axis is labeled “Loss” and has markings ranging from 0.4 to 1.1 in increments of 0.1 units. The graph shows 2 curves. The first curve for “Train” starts from (0, 1.1), slopes concave up, passing through coordinates (96.2, 0.68), (251, 0.47), and terminates at (400, 0.42). The second curve for “Valid” starts from (0, 1.1), slopes concave down, passing through coordinates (86.08, 0.68), (200, 0.46), and terminates at (400, 0.4). Note: All numerical data values are approximated.Offices NN models' accuracy and loss vs epochs curves. Source: Authors’ own work