Figure 4.

Analysis of absolute prediction error. Plot of target value vs. prediction error. Data was pooled into 20 bins according to their target values. For each bin, the mean absolute prediction error is plotted on the left y-axis. Then the number of values falling into the corresponding bin is shown with squares on the right y-axis. The lowest error is achieved for intermediate target values, the highest error occurs for low ones. The absolute error is not correlated to the number of values per bin. Thus, intensities within a certain range are more difficult to predict than others.