1.7 Assume you evenly split the data D into 3 disjoint subsets D1,D2, and D3. You run cross-validation to determine the better model between M1 and M2. When you achieve the following test errors: ErrD1( hM1)=0.32, ErrD2 ( hM1)=0.41, ErrD3( hM1)=0.15, ErrD1 ( hM2 )=0.24, ErrD2 ( hM2)=0.38, ErrD3(hM2)=0.17. Which model we should pick? choose one. M1 M2 Indifferent M2 could be better but may not be significantly