LP/MILP C API reference implementations. Use as reference when building new applications; do not edit in place. Build requires cuOpt installed (include and lib paths set).
\n| Example | \nType | \nDescription | \n
|---|---|---|
| lp_basic | \nLP | \nSimple LP: create problem, solve, get solution | \n
| lp_duals | \nLP | \nDual values and reduced costs | \n
| lp_warmstart | \nLP | \nPDLP warmstart (see README) | \n
| milp_basic | \nMILP | \nSimple MILP with integer variable | \n
| milp_production_planning | \nMILP | \nProduction planning with resource constraints | \n
| mps_solver | \nLP/MILP | \nSolve from MPS file via cuOptReadProblem | \n
Set include and library paths, then build and run.
\nUsing conda: Activate your cuOpt env first (conda activate cuopt), then:
# Paths from active conda env (CONDA_PREFIX is set when env is activated)\nexport INCLUDE_PATH=\"${CONDA_PREFIX}/include\"\nexport LIB_PATH=\"${CONDA_PREFIX}/lib\"\nexport LD_LIBRARY_PATH=\"${LIB_PATH}:${LD_LIBRARY_PATH}\"\n\n# Build and run (from this assets/ directory) — example: lp_basic\ngcc -I\"${INCLUDE_PATH}\" -L\"${LIB_PATH}\" -o lp_basic/lp_simple lp_basic/lp_simple.c -lcuopt\n./lp_basic/lp_simple\nFor the other examples, use the same pattern (e.g. lp_duals/lp_duals.c → lp_duals/lp_duals). mps_solver takes an MPS file path: ./mps_solver mps_solver/data/sample.mps.
Without conda, set INCLUDE_PATH and LIB_PATH to your cuOpt include and lib directories, then use the same gcc and LD_LIBRARY_PATH as above. Each subdirectory README has a one-line build/run for that example.