Check_dual: Quality Control for Post-processing
In yun-feng/WGDAP: Whole Genome Duplication Analysis Pipeline

Description Usage Arguments Details Value Examples

Check the dual problem of the lasso problem helps us to know how far we are from optimal.

1	Check_dual(data, par, res, g, lambda = c(8, 4))

`data`	The CNV dataset prepared for lasso (see `data_lasso`)
`par`	The parameter prepared for lasso (see `par_lasso`)
`res`	The result after solving lasso problem (see `Run_lasso`)
`g`	Integer-valued ploidy
`lambda`	L1-Penalty for all the variables (see `Run_lasso`)

This function would print out 5 values.

The first 2 are the dual variable and the 3rd and 4th should be your lambda parameters, which were used for Run_lasso. To achieve optimality, we should have the first 2 smaller or equal to the 3rd and 4th respectively. However, due to numeric error, sometimes it is slightly larger (less than 0.01

The last one is a duality gap, but it is a valid value only when the requirement above is satisfied. And it is strictly positive when it is valid. However, it should be a small value in most cases.

None