{
  "status": "PASS",
  "scope": "CPU environment and finite regression checks only; no main experiments run",
  "python_executable": "/root/userdata/liangyan/mdm_coverage_experiments/.venv/bin/python",
  "python_version": "3.13.13",
  "platform": "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
  "hostname": "cci-d40c5eb5-63fd-42ef-b8a1-3f3c712a44fa-0",
  "packages": {
    "numpy": "2.2.6",
    "scipy": "1.17.1",
    "matplotlib": "3.10.9"
  },
  "environment_mode": "project venv sharing existing Conda base site-packages",
  "visible_cpu_count": 192,
  "gpu_used": false,
  "library_thread_settings": {
    "OMP_NUM_THREADS": "1",
    "OPENBLAS_NUM_THREADS": "1",
    "MKL_NUM_THREADS": "1"
  },
  "binomial_mass_residual": 2.220446049250313e-16,
  "highs_randomization_smoke": {
    "objective": 0.0,
    "weights": [
      0.6666666666666666,
      0.33333333333333337
    ],
    "equality_residual": 0.0,
    "scope": "two-action LP only, not a general policy optimizer"
  },
  "headless_plotting": "PASS (canvas only; no publication figure generated)",
  "finite_regression": {
    "status": "PASS",
    "full_output_atoms": 1360,
    "posterior_heads": 104,
    "count_tv_identities": 16,
    "reverse_sensitivities": 50,
    "nonvacuous_bound_checks": 4,
    "scope": "finite examples only; not an all-policy test"
  },
  "finite_regression_sha256": "8eb778aa0276a33d9f43d31151cab7e9864924dd966743bec6d276cc000b5af6"
}