Table 4 Prediction results for \(^{13}C\) shifts using increasing numbers of spectra
| Â | Â | 100 | 250 | 500 | 1000 | 2500 | 5000 | 10000 | 25000 | 44370 |
|---|---|---|---|---|---|---|---|---|---|---|
2019 model | MAE (ppm) | 70.31 | 64.84 | 61.64 | 57.77 | 31.81 | 3.65 | 2.40 | 2.11 | 1.82 |
RMSE (ppm) | 83.30 | 79.86 | 76.36 | 70.71 | 36.27 | 5.41 | 3.35 | 4.09 | 3.13 | |
MASE | 1.49 | 1.39 | 1.32 | 1.23 | 0.67 | 0.07 | 0.05 | 0.04 | 0.04 | |
\(\sigma\) (ppm) | 52.29 | 52.51 | 52.69 | 48.15 | 28.05 | 6.64 | 5.08 | 5.13 | 4.57 | |
2023 model | MAE (ppm) | 24.8 | 21.45 | 22.77 | 17.11 | 15.0 | 11.18 | 9.63 | 8.21 | 7.65 |
RMSE (ppm) | 46.65 | 40.82 | 44.66 | 45.27 | 40.27 | 32.82 | 29.01 | 25.63 | 24.53 | |
MASE | 0.52 | 0.46 | 0.48 | 0.36 | 0.31 | 0.23 | 0.20 | 0.17 | 0.16 | |
\(\sigma\) (ppm) | 45.29 | 40.44 | 44.41 | 45.13 | 40.11 | 32.77 | 28.95 | 25.58 | 24.48 | |
HOSE code | MAE (ppm) | 20.81 | 18.99 | 17.68 | 18.85 | 16.14 | 15.2 | 14.02 | 12.14 | 10.98 |
RMSE (ppm) | 35.29 | 33.44 | 30.94 | 30.32 | 30.03 | 29.19 | 27.95 | 25.84 | 24.60 | |
\(\sigma\) (ppm) | 34.72 | 33.26 | 30.74 | 30.29 | 30.01 | 29.18 | 27.95 | 25.83 | 24.5 | |
MASE | 0.43 | 0.40 | 0.37 | 0.40 | 0.34 | 0.31 | 0.30 | 0.26 | 0.24 | |
Missing predictions | 17.6 | 19.6 | 26.3 | 33.9 | 41.4 | 51.0 | 57.9 | 76.2 | 85.2 |