{
 "returncode": 0,
 "stdout": "x_k5cFdt450.py:ladder_kernel at 1.90% error: 15 timed calls, each in a fresh process\n  review flag: 20,474 bytes, close to the 20,480-byte limit\n  NVIDIA A100-SXM4-80GB, torch 2.12.0+cu130, sandbox on\n  call  1/15: 403.132 ms, scorer's clock 407.711 ms\n  call  2/15: 402.963 ms, scorer's clock 406.736 ms\n  call  3/15: 402.863 ms, scorer's clock 406.761 ms\n  call  4/15: 402.905 ms, scorer's clock 408.212 ms\n  call  5/15: 404.030 ms, scorer's clock 411.040 ms\n  call  6/15: 405.473 ms, scorer's clock 409.284 ms\n  call  7/15: 403.951 ms, scorer's clock 408.772 ms\n  call  8/15: 404.879 ms, scorer's clock 409.432 ms\n  call  9/15: 405.528 ms, scorer's clock 409.418 ms\n  call 10/15: 402.959 ms, scorer's clock 407.095 ms\n  call 11/15: 402.806 ms, scorer's clock 406.684 ms\n  call 12/15: 402.970 ms, scorer's clock 407.637 ms\n  call 13/15: 402.881 ms, scorer's clock 406.920 ms\n  call 14/15: 402.823 ms, scorer's clock 406.372 ms\n  call 15/15: 402.825 ms, scorer's clock 406.542 ms\nMNIST 98.28% (108,108/110,000), 405.940 ms/call; hold-out (kmnist) 97.37%, 405.379 ms/call (floored by the scorer's clock); score 405.940 ms\nenergy: one more fresh process runs the method back to back on fresh MNIST draws for 20 s, between idle windows in which it is frozen\n  NVIDIA A100-SXM4-80GB: idle 68.5 W with the method frozen; telemetry reference 18.3 TFLOP/s at 7.73 J/TFLOP above idle (a healthy NVIDIA A100-SXM4-80GB reads 6-11)\n  the round trip alone: 13.3 mJ per call over 2,638 empty calls, subtracted\n  the method: 49 calls in 20.0 s, MNIST 98.28%, 403.364 ms per call, 226.9 W on average\nenergy 64776.719 mJ per call above idle\n",
 "stderr": "uards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:23:26.651000 1107 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:24:13.231000 1323 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:24:59.124000 1539 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:25:43.975000 1755 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:26:28.925000 1971 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:27:14.440000 2187 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:28:00.223000 2403 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:28:44.884000 2619 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:29:29.922000 2835 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:30:14.620000 3051 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\nW1020 12:31:24.535000 3268 site-packages/torch/_logging/_internal.py:1311] [1/0] Profiler record function <class 'torch.autograd.profiler.record_function'> will be ignored\n/usr/local/lib/python3.13/site-packages/torch/_dynamo/guards.py:1317: RuntimeWarning: Guards may run slower on Python 3.13.0. Consider upgrading to Python 3.13.1+.\n  warnings.warn(\n",
 "record": {
  "version": "mnist-a100/1.2.0 (protections of sutro-mnist-medium/3.0.0, tightened; energy column)",
  "file": "x_k5cFdt450.py",
  "function": "ladder_kernel",
  "source_sha256": "98e310b12031555088ee47e6f8170ad001fad1b6f0be28dcead493ff7682f5c3",
  "band_bp": 190,
  "sandboxed": true,
  "holdout": "kmnist",
  "device": "NVIDIA A100-SXM4-80GB",
  "torch": "2.12.0+cu130",
  "calls": [
   {
    "dataset": "kmnist",
    "ms": 403.1324157714844,
    "parent_ms": 410.5423270000017,
    "correct": 9753,
    "total": 10000,
    "overhead_ms": 2.830848002415628
   },
   {
    "dataset": "kmnist",
    "ms": 402.9634704589844,
    "parent_ms": 409.9130159999902,
    "correct": 9733,
    "total": 10000,
    "overhead_ms": 3.1771284968671694
   },
   {
    "dataset": "kmnist",
    "ms": 402.86309814453125,
    "parent_ms": 409.7096600000043,
    "correct": 9731,
    "total": 10000,
    "overhead_ms": 2.9488115017124983
   },
   {
    "dataset": "mnist",
    "ms": 402.90509033203125,
    "parent_ms": 410.67812000000004,
    "correct": 9828,
    "total": 10000,
    "overhead_ms": 2.465818997230599
   },
   {
    "dataset": "mnist",
    "ms": 404.03045654296875,
    "parent_ms": 414.03973399999927,
    "correct": 9853,
    "total": 10000,
    "overhead_ms": 2.9998344988669032
   },
   {
    "dataset": "mnist",
    "ms": 405.4732666015625,
    "parent_ms": 412.17317300004197,
    "correct": 9816,
    "total": 10000,
    "overhead_ms": 2.8888234994660706
   },
   {
    "dataset": "kmnist",
    "ms": 403.9505920410156,
    "parent_ms": 411.8650789999947,
    "correct": 9731,
    "total": 10000,
    "overhead_ms": 3.092594000074314
   },
   {
    "dataset": "mnist",
    "ms": 404.8793640136719,
    "parent_ms": 412.4472100000389,
    "correct": 9823,
    "total": 10000,
    "overhead_ms": 3.0153230025894118
   },
   {
    "dataset": "mnist",
    "ms": 405.5275573730469,
    "parent_ms": 411.9692540000415,
    "correct": 9808,
    "total": 10000,
    "overhead_ms": 2.551045499102429
   },
   {
    "dataset": "mnist",
    "ms": 402.9593505859375,
    "parent_ms": 409.73294700000906,
    "correct": 9833,
    "total": 10000,
    "overhead_ms": 2.6380825024948535
   },
   {
    "dataset": "mnist",
    "ms": 402.8057556152344,
    "parent_ms": 409.19028000001845,
    "correct": 9829,
    "total": 10000,
    "overhead_ms": 2.5060495024010834
   },
   {
    "dataset": "mnist",
    "ms": 402.9696044921875,
    "parent_ms": 410.2343550000569,
    "correct": 9822,
    "total": 10000,
    "overhead_ms": 2.59713699969916
   },
   {
    "dataset": "mnist",
    "ms": 402.8805236816406,
    "parent_ms": 409.5254600000544,
    "correct": 9831,
    "total": 10000,
    "overhead_ms": 2.6051945015829006
   },
   {
    "dataset": "mnist",
    "ms": 402.82318115234375,
    "parent_ms": 409.51269100003174,
    "correct": 9857,
    "total": 10000,
    "overhead_ms": 3.1404360017313593
   },
   {
    "dataset": "mnist",
    "ms": 402.8252258300781,
    "parent_ms": 409.4960139999557,
    "correct": 9808,
    "total": 10000,
    "overhead_ms": 2.9542479975129936
   }
  ],
  "problems": [],
  "summary": "MNIST 98.28% (108,108/110,000), 405.940 ms/call; hold-out (kmnist) 97.37%, 405.379 ms/call (floored by the scorer's clock)",
  "ranked_ms": 405.93971346513325,
  "energy": {
   "mj_per_call": 64776.718850052814,
   "problems": [],
   "net_mj_before_control": 64789.98309483402,
   "control_mj_per_call": 13.264244781202132,
   "control_calls": 2638,
   "gross_mj_per_call": 92807.16326530612,
   "idle_w": 68.49933163864188,
   "active_w": 226.90465693802415,
   "calls": 49,
   "seconds": 20.041682093999952,
   "ms_median": 403.36383056640625,
   "timed_mnist_ms_median": 402.9593505859375,
   "correct": 481595,
   "total": 490000,
   "reference": {
    "tflops_per_s": 18.323070531490064,
    "j_per_tflop": 7.734880072257791,
    "exact": true,
    "band": [
     [
      6.0,
      11.0
     ],
     [
      15.0,
      23.0
     ]
    ]
   },
   "device": {
    "name": "NVIDIA A100-SXM4-80GB",
    "uuid": "GPU-756db17f-4fa6-42b6-ee39-e4598d59ad18",
    "vbios": "92.00.9E.00.01",
    "driver": "610.57.04",
    "power_limit_w": 400.0
   },
   "windows": [
    {
     "name": "idle 1",
     "seconds": 5.011123072000032,
     "joules": 331.883,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "reference",
     "seconds": 5.220604890000004,
     "joules": 1090.276,
     "matmuls": 696,
     "dim": 4096,
     "exact": true
    },
    {
     "name": "idle 2",
     "seconds": 5.004787133000036,
     "joules": 340.321,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "control",
     "seconds": 5.004302798000026,
     "joules": 375.087,
     "calls": 2638
    },
    {
     "name": "idle 3",
     "seconds": 5.0059255510000185,
     "joules": 340.014,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "method",
     "seconds": 20.041682093999952,
     "joules": 4547.551,
     "calls": 49,
     "ms": [
      434.34906005859375,
      403.4170837402344,
      403.53485107421875,
      403.2593994140625,
      403.5696716308594,
      403.5328063964844,
      403.1733703613281,
      403.388427734375,
      403.6792297363281,
      403.4396057128906,
      403.4437255859375,
      403.4129943847656,
      402.9757385253906,
      404.0007629394531,
      403.13958740234375,
      403.0771179199219,
      403.04638671875,
      403.40069580078125,
      403.3986511230469,
      403.4078674316406,
      403.4314270019531,
      403.36895751953125,
      403.2051086425781,
      403.3843078613281,
      403.3034362792969,
      403.52972412109375,
      403.3576965332031,
      403.5491943359375,
      403.5072021484375,
      403.49285888671875,
      403.36383056640625,
      403.857421875,
      403.26348876953125,
      403.198974609375,
      403.25018310546875,
      403.5788879394531,
      403.39251708984375,
      403.1600646972656,
      403.11090087890625,
      403.1866760253906,
      403.26348876953125,
      403.1877136230469,
      403.2194519042969,
      403.24609375,
      403.17132568359375,
      403.3013610839844,
      403.1539306640625,
      403.1856689453125,
      403.167236328125
     ],
     "correct": [
      9848,
      9813,
      9814,
      9848,
      9830,
      9814,
      9847,
      9822,
      9836,
      9792,
      9834,
      9848,
      9819,
      9838,
      9830,
      9799,
      9843,
      9791,
      9817,
      9845,
      9821,
      9855,
      9838,
      9825,
      9841,
      9824,
      9837,
      9842,
      9841,
      9845,
      9828,
      9835,
      9848,
      9819,
      9802,
      9838,
      9829,
      9806,
      9840,
      9823,
      9829,
      9809,
      9834,
      9835,
      9814,
      9827,
      9829,
      9810,
      9843
     ]
    },
    {
     "name": "idle 4",
     "seconds": 5.005475194999917,
     "joules": 345.76,
     "utilization_max": 0,
     "contexts": 1
    }
   ]
  }
 }
}
