Table 1 Mean absolute error (MAE) values, as well as the standard error of the mean (σM; shown in brackets), obtained using the different ML algorithms following prediction of fractional drug release during nested cross-validation (i.e., n = 10 trials)
From: Machine learning models to accelerate the design of polymeric long-acting injectables
LGBM | RF | NGB | XGB | DT | NN | k-NN | SVR | lasso | PLS | MLR | |
|---|---|---|---|---|---|---|---|---|---|---|---|
The inner loop (validation) | 0.125 (0.039) | 0.130 (0.041) | 0.128 (0.040) | 0.133 (0.042) | 0.154 (0.049) | 0.166 (0.053) | 0.179 (0.057) | 0.206 (0.065) | 0.242 (0.076) | 0.246 (0.078) | 0.244 (0.081) |
Outer loop (test) | 0.114 (0.036) | 0.119 (0.038) | 0.122 (0.039) | 0.126 (0.040) | 0.154 (0.049) | 0.184 (0.058) | 0.176 (0.056) | 0.248 (0.079) | 0.255 (0.081) | 0.252 (0.080) | 0.255 (0.081) |
