Table 1
Comparison of age and metallicity prediction among different models.
| MSE | MAE | Bias(SD) | ||||
|---|---|---|---|---|---|---|
| Age | Metallicity | Age | Metallicity | Age | Metallicity | |
| Image-only model | 0.0441 | 0.0272 | 0.1684 | 0.1302 | 0.0426(0.2057) | 0.0122(0.1644) |
| Spectra-only model | 0.0197 | 0.0161 | 0.1129 | 0.1024 | 0.0365(0.1355) | -0.0056(0.1269) |
| MMLforGalAM | 0.0227 | 0.0197 | 0.1215 | 0.1121 | 0.0034(0.1506) | -0.0091(0.1402) |
| MMLforGalAMwithReal | 0.0168 | 0.0163 | 0.1058 | 0.1041 | -0.0088(0.1293) | -0.0175(0.1266) |
Notes. The evaluation metrics are described in Sect. 4.1. The image-only model predicts age and metallicity based only on images; the spectra-only model uses spectra only. Our MMLforGalAM model employs simulated spectral features generated by ℳ2 and integrates them with image features. The MMLforGalAMwithReal model is trained on real spectral features extracted by ℳ1, which are directly input to ℳ4, bypassing ℳ2.
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.