σ_MAD (Eq. 11), the slopes of the < Δ_z > − z_SPec piecewise linear fit (Eq. 9) and the total variation distance between the z_Ph0t0 and z_SPec sample distributions (Eq. 10) as a function of the balancing threshold that defines the nearly balanced training subset for steps 2 and 3 of our methods. As the threshold approaches the maximum number density (2,626 in this case), the results from step 2 are expected to converge to those from step 1, and those from step 3 are expected to converge to those from step 3\2 in which step 2 is skipped. The dotted curves show the total variation distances for the simulated z_photo sample distributions that are expected to have no mode collapse. We compare the methods using the SDSS data and the network Net_P.