True redshift distributions of the training samples used to run DNF in all six cases. The training samples include magnitudes brighter than m_VIS = 24.5. The true redshift comes from the Flagship simulation. The four training samples with almost identical true redshift distributions have the same completeness drop in m_VIS and only differ in the photometric quality. The numbers of training objects for the six samples are about 3.4 ⋅ 10⁵, 1.8 ⋅ 10⁵, 1.8 ⋅ 10⁵, 1.8 ⋅ 10⁵, 1.8 ⋅ 10⁵, and 8.4 ⋅ 10⁴ from top to bottom labels in the legend, respectively.