Table 1.
Comparisons of parameter estimates for different catalogs using various methodologies.
| Catalog | Band | Method | Bias | σNMAD | η* [%] | tproc [h/CPU/1000] |
|---|---|---|---|---|---|---|
| JWST/PRIMER | F090W, ...F770W (9) | SOM25 | –0.03 | –0.04 | 0.12 | 0.18 | 0.7 | 7.3 | ∼0.001 |
| JWST/PRIMER | F090W, ...F770W (9) | SHAPE25 | –0.03 | –0.05 | 0.16 | 0.18 | 2.8 | 7.3 | |
| COSMOS2020/full-phot | u, g, ...ch2 (12) | SHAPE30 | –0.03 | 0.01 | 0.19 | 0.25 | 3.1 | 11.6 | |
| COSMOS2020/miss-phot | u, g, ... H (9) | SHAPE30 | –0.02 | –0.01 | 0.20 | 0.25 | 3.5 | 11.2 | |
| COSMOS2020/miss-phot | u, g, ... H (9) | BAGPIPES | 0.19 | 0.18 | 0.22 | 0.35 | 19.2 | 14.1 | ∼10 |
Notes. Three key statistical metrics for the derived parameters are reported in the order of M★ and SFR. For a fair comparison, the outlier fraction η* is defined as the fraction of objects with |logXSOM − logXSED|> 0.5 dex. The computational efficiency (tproc) of each model inevitably depends on factors such as CPU architecture and the complexity of the SED fitting templates, thus, we provide only an order-of-magnitude estimate as a reference.
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.