{
 "returncode": 0,
 "stdout": "a_k2t_600.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: 628.312 ms, scorer's clock 630.723 ms\n  call  2/15: 627.548 ms, scorer's clock 630.314 ms\n  call  3/15: 628.555 ms, scorer's clock 632.049 ms\n  call  4/15: 632.239 ms, scorer's clock 635.811 ms\n  call  5/15: 628.678 ms, scorer's clock 632.800 ms\n  call  6/15: 628.495 ms, scorer's clock 631.256 ms\n  call  7/15: 629.558 ms, scorer's clock 632.710 ms\n  call  8/15: 629.015 ms, scorer's clock 631.591 ms\n  call  9/15: 629.111 ms, scorer's clock 632.258 ms\n  call 10/15: 628.333 ms, scorer's clock 631.687 ms\n  call 11/15: 632.745 ms, scorer's clock 635.839 ms\n  call 12/15: 628.488 ms, scorer's clock 630.464 ms\n  call 13/15: 633.013 ms, scorer's clock 635.651 ms\n  call 14/15: 628.623 ms, scorer's clock 631.225 ms\n  call 15/15: 628.396 ms, scorer's clock 630.715 ms\nMNIST 98.36% (108,200/110,000), 628.957 ms/call; hold-out (kmnist) 97.34%, 630.646 ms/call; score 630.646 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 83.9 W with the method frozen; telemetry reference 18.4 TFLOP/s at 8.46 J/TFLOP above idle (a healthy NVIDIA A100-SXM4-80GB reads 6-11)\n  the round trip alone: 11.5 mJ per call over 3,192 empty calls, subtracted\n  the method: 32 calls in 20.2 s, MNIST 98.30%, 628.390 ms per call, 272.8 W on average\nenergy 119454.127 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(\nW1019 22:14:20.917000 1122 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(\nW1019 22:15:00.729000 1341 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(\nW1019 22:15:40.399000 1560 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(\nW1019 22:16:19.118000 1779 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(\nW1019 22:16:58.109000 1998 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(\nW1019 22:17:35.680000 2217 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(\nW1019 22:18:14.434000 2436 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(\nW1019 22:18:52.116000 2655 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(\nW1019 22:19:29.985000 2874 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(\nW1019 22:20:06.473000 3093 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(\nW1019 22:21:10.428000 3313 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": "a_k2t_600.py",
  "function": "ladder_kernel",
  "source_sha256": "cc423b95ed940a5bcd5dc3b08f71502dddbcee3b42cbc34fc5322819378aef2a",
  "band_bp": 190,
  "sandboxed": true,
  "holdout": "kmnist",
  "device": "NVIDIA A100-SXM4-80GB",
  "torch": "2.12.0+cu130",
  "calls": [
   {
    "dataset": "mnist",
    "ms": 628.3115234375,
    "parent_ms": 633.2819490000006,
    "correct": 9833,
    "total": 10000,
    "overhead_ms": 2.5585969999320497
   },
   {
    "dataset": "mnist",
    "ms": 627.54833984375,
    "parent_ms": 632.7865660000072,
    "correct": 9811,
    "total": 10000,
    "overhead_ms": 2.472660003454422
   },
   {
    "dataset": "mnist",
    "ms": 628.55517578125,
    "parent_ms": 634.3840830000005,
    "correct": 9835,
    "total": 10000,
    "overhead_ms": 2.3347455008615725
   },
   {
    "dataset": "kmnist",
    "ms": 632.2388305664062,
    "parent_ms": 638.301848999987,
    "correct": 9729,
    "total": 10000,
    "overhead_ms": 2.4909410021035114
   },
   {
    "dataset": "mnist",
    "ms": 628.6781616210938,
    "parent_ms": 635.2941170000008,
    "correct": 9826,
    "total": 10000,
    "overhead_ms": 2.4939339996245735
   },
   {
    "dataset": "mnist",
    "ms": 628.4951782226562,
    "parent_ms": 633.8149430000044,
    "correct": 9850,
    "total": 10000,
    "overhead_ms": 2.5594179993179296
   },
   {
    "dataset": "mnist",
    "ms": 629.5580444335938,
    "parent_ms": 635.2586809999821,
    "correct": 9829,
    "total": 10000,
    "overhead_ms": 2.548944998687716
   },
   {
    "dataset": "mnist",
    "ms": 629.0152587890625,
    "parent_ms": 634.3842370000061,
    "correct": 9848,
    "total": 10000,
    "overhead_ms": 2.793331998793292
   },
   {
    "dataset": "kmnist",
    "ms": 629.1112670898438,
    "parent_ms": 634.7556930000451,
    "correct": 9733,
    "total": 10000,
    "overhead_ms": 2.4972120009747414
   },
   {
    "dataset": "mnist",
    "ms": 628.3330078125,
    "parent_ms": 634.186399999976,
    "correct": 9825,
    "total": 10000,
    "overhead_ms": 2.4994424999390503
   },
   {
    "dataset": "kmnist",
    "ms": 632.744873046875,
    "parent_ms": 638.1456820000153,
    "correct": 9747,
    "total": 10000,
    "overhead_ms": 2.306603499034736
   },
   {
    "dataset": "kmnist",
    "ms": 628.4876708984375,
    "parent_ms": 633.3775909999986,
    "correct": 9727,
    "total": 10000,
    "overhead_ms": 2.913805498262491
   },
   {
    "dataset": "mnist",
    "ms": 633.0130615234375,
    "parent_ms": 638.2672230000139,
    "correct": 9829,
    "total": 10000,
    "overhead_ms": 2.6158089954224124
   },
   {
    "dataset": "mnist",
    "ms": 628.6231079101562,
    "parent_ms": 633.7769130000197,
    "correct": 9844,
    "total": 10000,
    "overhead_ms": 2.5524064996156994
   },
   {
    "dataset": "mnist",
    "ms": 628.3960571289062,
    "parent_ms": 633.309044999919,
    "correct": 9870,
    "total": 10000,
    "overhead_ms": 2.593644500028404
   }
  ],
  "problems": [],
  "summary": "MNIST 98.36% (108,200/110,000), 628.957 ms/call; hold-out (kmnist) 97.34%, 630.646 ms/call",
  "ranked_ms": 630.6456604003906,
  "energy": {
   "mj_per_call": 119454.1269780938,
   "problems": [],
   "net_mj_before_control": 119465.67480614973,
   "control_mj_per_call": 11.547828055936455,
   "control_calls": 3192,
   "gross_mj_per_call": 172519.46875,
   "idle_w": 83.88070328151328,
   "active_w": 272.76191376282276,
   "calls": 32,
   "seconds": 20.239713542999993,
   "ms_median": 628.3895874023438,
   "timed_mnist_ms_median": 628.55517578125,
   "correct": 314571,
   "total": 320000,
   "reference": {
    "tflops_per_s": 18.426841070122492,
    "j_per_tflop": 8.464939016538759,
    "exact": true,
    "band": [
     [
      6.0,
      11.0
     ],
     [
      15.0,
      23.0
     ]
    ]
   },
   "device": {
    "name": "NVIDIA A100-SXM4-80GB",
    "uuid": "GPU-f779a8e7-f7c9-d9a6-865b-fca07311dcb9",
    "vbios": "92.00.94.00.04",
    "driver": "610.57.04",
    "power_limit_w": 400.0
   },
   "windows": [
    {
     "name": "idle 1",
     "seconds": 5.005926870000053,
     "joules": 421.457,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "reference",
     "seconds": 5.1912051150000025,
     "joules": 1243.22,
     "matmuls": 696,
     "dim": 4096,
     "exact": true
    },
    {
     "name": "idle 2",
     "seconds": 5.0051885359999915,
     "joules": 414.509,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "control",
     "seconds": 5.0037773990000005,
     "joules": 454.788,
     "calls": 3192
    },
    {
     "name": "idle 3",
     "seconds": 5.005148304000045,
     "joules": 421.578,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "method",
     "seconds": 20.239713542999993,
     "joules": 5520.623,
     "calls": 32,
     "ms": [
      654.8701782226562,
      628.3475341796875,
      628.22265625,
      628.1448364257812,
      628.6166381835938,
      628.2841796875,
      628.3413696289062,
      628.5786743164062,
      628.4908447265625,
      628.5221557617188,
      628.1481323242188,
      628.0432739257812,
      628.6468505859375,
      628.2186279296875,
      628.4812622070312,
      628.6605224609375,
      628.4473876953125,
      628.0322265625,
      628.3856811523438,
      628.3934936523438,
      628.4331665039062,
      628.2138671875,
      628.5542602539062,
      628.8831787109375,
      628.3151245117188,
      628.2843627929688,
      628.4327392578125,
      628.051025390625,
      628.1024780273438,
      628.7613525390625,
      628.10302734375,
      628.4620971679688
     ],
     "correct": [
      9827,
      9848,
      9851,
      9844,
      9836,
      9820,
      9823,
      9830,
      9839,
      9847,
      9812,
      9822,
      9834,
      9819,
      9821,
      9814,
      9821,
      9830,
      9825,
      9833,
      9824,
      9837,
      9827,
      9817,
      9848,
      9821,
      9836,
      9831,
      9831,
      9824,
      9832,
      9847
     ]
    },
    {
     "name": "idle 4",
     "seconds": 5.0043854929999725,
     "joules": 418.029,
     "utilization_max": 0,
     "contexts": 1
    }
   ]
  }
 }
}
