inst/tinytest/test_random_lp.R

In scs: Splitting Conic Solver

## A Random LP Problem

set.seed(1)
m  <-  100
n  <- 100
A <- matrix(rnorm(m * n), m, n)
b  <-  rnorm(m)
c  <-  rnorm(n)
cone  <-  list(z = m)
params  <-  list(eps_rel = 1e-4, eps_abs = 1e-4, max_iters = 5000)
sol <- scs(A = A, b = b, obj = c, cone = cone, control = params)
expect_equal(sol$info$status, "solved")
expect_equal(sol$info$pobj, 13.5089, tolerance = 1e-4)
expect_equal(sol$info$dobj, 13.5089, tolerance = 1e-4)