Open Access

Table 2.

Description of the data refinement test.

Data refinement tests Description Training set Validation set
Baseline Galaxies without measurements in any of the bands are excluded. W1+W3: 22 378 W1+W3: 5581
G09+COSMOS: 9073 G09+COSMOS: 2269
Cross-validation (CV) Comparison of i-band filters to confirm that the two CFHT filters provide consistent and comparable measurements. W1+W3: 23 592 W1+W3: 5581
G09+COSMOS: 7606 G09+COSMOS: 1902
Impute (non-NaN) Investigation into handling galaxies with NaN values in the NB fluxes and the use of interpolation through a quadratic fit. W1+W3: 30 383 W1+W3: 7561
G09+COSMOS: 7606 G09+COSMOS: 1902
Low signal-to-noise ratio fluxes (Low S/N) Investigation of the treatment of low signal-to-noise ratio fluxes including those sources with magnitudes equal to 99. W1+W3: 22 118 W1+W3: 5529
G09+COSMOS: 7606 G09+COSMOS: 1902
Bright galaxies (i < 22.5) We limit the magnitudes of the selected band to values less than 22.5. W1+W3: 22 378 W1+W3: 5581
G09+COSMOS: 7606 G09+COSMOS: 1902
Galactic extinction correction (Ext. Correction) Study on the Galactic extinction correction in the BB and NB fluxes. W1+W3: 22 378 W1+W3: 5581
G09+COSMOS: 7606 G09+COSMOS: 1902
Only NBs (NBs) Analysis of the importance of the more general features of the SED provided in the BBs. W1+W3: 22 378 W1+W3: 5581
G09+COSMOS: 7606 G09+COSMOS: 1902
SED in the blue wavelength range (blue-lines) Indicates the importance of the lines detected in the blue range. W1+W3: 22 378 W1+W3: 5581
G09+COSMOS: 7606 G09+COSMOS: 1902

Notes. Summary of studies on observational effects in photometric redshift estimation. Dimensions of the training and validation sets for the combination of data in W1 and W3, and for the combination of data in the fields G09 and COSMOS.

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.