Residual norm, norm of normal equations, and matrix norm for regressor_linear_ solve.
    0 KSP resid norm 1.516575088810e+00
    1 KSP resid norm 7.758996082099e-01 normal eq resid norm 8.408142938043e-01 matrix norm 1.228473174906e+00
    2 KSP resid norm 3.380617018914e-01 normal eq resid norm 2.920128359737e-16 matrix norm 1.837117307087e+00
PetscRegressor Object: 1 MPI process
  type: linear
    PetscRegressor Linear Type: ols
    KSP Object: (regressor_linear_) 1 MPI process
      type: lsqr
        standard error not computed
        using inexact matrix norm
      maximum iterations=10000, initial guess is zero
      tolerances: relative=1e-05, absolute=1e-50, divergence=10000.
      left preconditioning
      using UNPRECONDITIONED norm type for convergence test
    PC Object: (regressor_linear_) 1 MPI process
      type: none
      linear system matrix, followed by the matrix used to construct the preconditioner:
      Mat Object: 1 MPI process
        type: composite
        rows=5, cols=2
      Mat Object: 1 MPI process
        type: normal
        rows=2, cols=2
    total KSP iterations: 2
    Intercept=0.0857143
Intercept is 0.085714
Coefficients are
Vec Object: 1 MPI process
  type: seq
0.4
1.42857
Predicted values are
Vec Object: 1 MPI process
  type: seq
1.11429
0.242857
0.0857143
0.642857
1.91429