Schematic of SALT-II training test pipeline. From an input SN Ia model and cosmology, N sets of SN Ia training data and a single test set of SN Ia light curves are generated (Sections 3, 4.2, and 4.3). Each set of training data is used to constrain a new SALT-II model (Section 2). Then, each new SALT-II model is used to determine test set SN stretch and color. From this information, SALT-II model parameters α, β, and M₀, and test set SN distances μ_fit are determined for each trained model (Section 4.5). The recovered test set stretch and color parameter distributions are used in conjunction with the trained models to generate and fit one bias correction set for each model (Section 4.6). Using the bias corrections, corrected test set distances μ are calculated for each model. The resulting HDs are fitted to obtain N sets of best-fit cosmology parameters w and Ω_M.