Commit b311a99
authored
ci(thirdparty-pulp): skip test_numpy_float in nightly cuOpt run (NVIDIA#1190)
## Summary
PuLP's `test_numpy_float` calls `model.solve()` without an explicit solver and relies on PuLP's default-solver auto-detection. `CUOPT` is not on that default list, so in our cuopt-only nightly test environment the call raises `PulpError: No solver available` and the test fails for reasons unrelated to cuOpt.
This PR deselects the test in `ci/thirdparty-testing/run_pulp_tests.sh` with an inline comment explaining the cause.
## Context — other cuOpt-related PuLP failures
| Test | Status |
|---|---|
| `test_unbounded` | Fixed upstream by [coin-or/pulp#901](coin-or/pulp#901) (tracked in NVIDIA#1114) |
| `test_integer_infeasible_2` | Fixed upstream by [coin-or/pulp#901](coin-or/pulp#901) (tracked in NVIDIA#1114) |
| `test_infeasible_2` | Real cuOpt bug — PDLP returns Optimal for an infeasible LP. Tracked in NVIDIA#1115. Not skipped here. |
| `test_numpy_float` | **This PR** |
## Test plan
- [ ] Nightly third-party-PuLP job no longer reports `test_numpy_float` as failing
- [ ] Other cuOpt PuLP tests still execute (no over-broad deselect)
Authors:
- Ramakrishnap (https://github.com/rgsl888prabhu)
Approvers:
- Miles Lubin (https://github.com/mlubin)
- Trevor McKay (https://github.com/tmckayus)
URL: NVIDIA#11901 parent f3ab099 commit b311a99
1 file changed
Lines changed: 4 additions & 0 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
30 | 30 | | |
31 | 31 | | |
32 | 32 | | |
| 33 | + | |
| 34 | + | |
| 35 | + | |
33 | 36 | | |
34 | 37 | | |
35 | 38 | | |
36 | 39 | | |
37 | 40 | | |
| 41 | + | |
38 | 42 | | |
39 | 43 | | |
40 | 44 | | |
| |||
0 commit comments