Fig. 5.

Learning curves generated by using dGBDT for multiple data splits of each of the datasets in Table 1. a The entire set of learning curve scores, $L{C}_{\mathit{raw}}$, where each data point is the mean absolute error of predictions computed on test set E as a function of the training set size $m_k$. A subset of scores in which the sample size is above $m_{\mathit{kmin}}$ (dashed black line) was considered for curve fitting. b Three curves were generated to represent the fit: $q_{0.1}$ (blue curve) and $q_{0.9}$ (green curve) representing the variability of the fit, and $\tilde{y}$ (black curve) representing the learning curve fit