Table 3
Error metric for five pairs of randomly selected planets.
| Planets | Error metric | DNN | XG |
|---|---|---|---|
| F1800 | R2 | 0.9971 | 0.9971 |
| K800 | RMSE | 48.7536 | 49.1047 |
| G1800 | R2 | 0.9817 | 0.9813 |
| K1800 | RMSE | 117.3391 | 118.8262 |
| F2000 | R2 | 0.9835 | 0.9843 |
| G2200 | RMSE | 121.7374 | 118.7824 |
| G2000 | R2 | 0.9931 | 0.9953 |
| K600 | RMSE | 77.0609 | 63.4323 |
| F400 | R2 | 0.9975 | 0.9987 |
| F1600 | RMSE | 44.6912 | 32.3724 |
Notes. R2 and RMSE for Tgas data of the five independent pairs of randomly selected planets removed from training (Fig. 9, Figs. B.5–B.8). All of the R2 values are very close to one, showing good agreement between the prediction and the ground truth. Although the RMSE is high due to large absolute deviations, the ML models successfully capture the underlying trends and variability, as evidenced by the high R2 values.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.