{
 "returncode": 0,
 "stdout": "mnist_d4_ladder_kernel.py:ladder_kernel at 2.30% error: 15 timed calls, each in a fresh process\n  review flag: 20,451 bytes, close to the 20,480-byte limit\n  NVIDIA A100-SXM4-80GB, torch 2.12.0+cu130, sandbox on\n  call  1/15: 581.867 ms, scorer's clock 586.384 ms\n  call  2/15: 580.703 ms, scorer's clock 584.340 ms\n  call  3/15: 580.325 ms, scorer's clock 581.630 ms\n  call  4/15: 579.105 ms, scorer's clock 582.243 ms\n  call  5/15: 577.236 ms, scorer's clock 580.625 ms\n  call  6/15: 578.979 ms, scorer's clock 582.368 ms\n  call  7/15: 580.077 ms, scorer's clock 583.113 ms\n  call  8/15: 580.091 ms, scorer's clock 588.868 ms\n  call  9/15: 581.160 ms, scorer's clock 584.431 ms\n  call 10/15: 578.801 ms, scorer's clock 582.345 ms\n  call 11/15: 583.232 ms, scorer's clock 586.053 ms\n  call 12/15: 579.437 ms, scorer's clock 583.989 ms\n  call 13/15: 578.386 ms, scorer's clock 582.683 ms\n  call 14/15: 579.453 ms, scorer's clock 583.917 ms\n  call 15/15: 579.984 ms, scorer's clock 585.108 ms\nMNIST 97.86% (107,646/110,000), 581.317 ms/call; hold-out (fashion) 88.67%, 579.653 ms/call (floored by the scorer's clock); score 581.317 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 72.9 W with the method frozen; telemetry reference 18.2 TFLOP/s at 8.14 J/TFLOP above idle (a healthy NVIDIA A100-SXM4-80GB reads 6-11)\n  the round trip alone: 13.7 mJ per call over 2,393 empty calls, subtracted\n  the method: 35 calls in 20.5 s, MNIST 97.87%, 579.805 ms per call, 182.6 W on average\nenergy 64350.464 mJ per call above idle\n",
 "stderr": "/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n/usr/local/lib/python3.13/site-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.\n  warnings.warn(\n",
 "record": {
  "version": "mnist-a100/1.2.0 (protections of sutro-mnist-medium/3.0.0, tightened; energy column)",
  "file": "mnist_d4_ladder_kernel.py",
  "function": "ladder_kernel",
  "source_sha256": "f4c69bf486197cd00721320a90dbd644ad734b3371e65ea2fff498d9304ebbdc",
  "band_bp": 230,
  "sandboxed": true,
  "holdout": "fashion",
  "device": "NVIDIA A100-SXM4-80GB",
  "torch": "2.12.0+cu130",
  "calls": [
   {
    "dataset": "mnist",
    "ms": 581.8665161132812,
    "parent_ms": 590.9030680000029,
    "correct": 9796,
    "total": 10000,
    "overhead_ms": 4.519539498946358
   },
   {
    "dataset": "mnist",
    "ms": 580.7032470703125,
    "parent_ms": 588.8281269999993,
    "correct": 9789,
    "total": 10000,
    "overhead_ms": 4.4883084952995205
   },
   {
    "dataset": "fashion",
    "ms": 580.3253784179688,
    "parent_ms": 585.8928109999795,
    "correct": 8853,
    "total": 10000,
    "overhead_ms": 4.2631239986957326
   },
   {
    "dataset": "mnist",
    "ms": 579.1047973632812,
    "parent_ms": 584.4768489999979,
    "correct": 9805,
    "total": 10000,
    "overhead_ms": 2.2340215009348867
   },
   {
    "dataset": "fashion",
    "ms": 577.2359619140625,
    "parent_ms": 582.6250510000079,
    "correct": 8860,
    "total": 10000,
    "overhead_ms": 1.9998429990453133
   },
   {
    "dataset": "mnist",
    "ms": 578.9788208007812,
    "parent_ms": 584.4246679999969,
    "correct": 9783,
    "total": 10000,
    "overhead_ms": 2.057150999908572
   },
   {
    "dataset": "mnist",
    "ms": 580.0765380859375,
    "parent_ms": 585.0376689999734,
    "correct": 9762,
    "total": 10000,
    "overhead_ms": 1.9246759994196054
   },
   {
    "dataset": "mnist",
    "ms": 580.0908813476562,
    "parent_ms": 591.1534389999815,
    "correct": 9780,
    "total": 10000,
    "overhead_ms": 2.2851294996808065
   },
   {
    "dataset": "fashion",
    "ms": 581.159912109375,
    "parent_ms": 589.5111379999776,
    "correct": 8876,
    "total": 10000,
    "overhead_ms": 5.0805260014547
   },
   {
    "dataset": "mnist",
    "ms": 578.8006591796875,
    "parent_ms": 587.2467280000251,
    "correct": 9787,
    "total": 10000,
    "overhead_ms": 4.901508502132401
   },
   {
    "dataset": "mnist",
    "ms": 583.2324829101562,
    "parent_ms": 591.1610800000062,
    "correct": 9800,
    "total": 10000,
    "overhead_ms": 5.107886001660972
   },
   {
    "dataset": "mnist",
    "ms": 579.4365234375,
    "parent_ms": 588.5117039999841,
    "correct": 9761,
    "total": 10000,
    "overhead_ms": 4.522368001558789
   },
   {
    "dataset": "mnist",
    "ms": 578.3859252929688,
    "parent_ms": 587.0877380000366,
    "correct": 9789,
    "total": 10000,
    "overhead_ms": 4.404621498279084
   },
   {
    "dataset": "fashion",
    "ms": 579.4529418945312,
    "parent_ms": 588.7640140000485,
    "correct": 8878,
    "total": 10000,
    "overhead_ms": 4.846847498616171
   },
   {
    "dataset": "mnist",
    "ms": 579.984375,
    "parent_ms": 589.2183839999916,
    "correct": 9794,
    "total": 10000,
    "overhead_ms": 4.10988500106123
   }
  ],
  "problems": [],
  "summary": "MNIST 97.86% (107,646/110,000), 581.317 ms/call; hold-out (fashion) 88.67%, 579.653 ms/call (floored by the scorer's clock)",
  "ranked_ms": 581.3173686516459,
  "energy": {
   "mj_per_call": 64350.4638285023,
   "problems": [],
   "net_mj_before_control": 64364.15677154723,
   "control_mj_per_call": 13.692943044921504,
   "control_calls": 2393,
   "gross_mj_per_call": 107118.6,
   "idle_w": 72.88440982256267,
   "active_w": 182.60735849843738,
   "calls": 35,
   "seconds": 20.531215340000017,
   "ms_median": 579.80517578125,
   "timed_mnist_ms_median": 579.984375,
   "correct": 342534,
   "total": 350000,
   "reference": {
    "tflops_per_s": 18.222718596043876,
    "j_per_tflop": 8.142335826424192,
    "exact": true,
    "band": [
     [
      6.0,
      11.0
     ],
     [
      15.0,
      23.0
     ]
    ]
   },
   "device": {
    "name": "NVIDIA A100-SXM4-80GB",
    "uuid": "GPU-9344a61d-da46-751c-4941-68fc9620cfbc",
    "vbios": "92.00.9E.00.02",
    "driver": "580.95.05",
    "power_limit_w": 500.0
   },
   "windows": [
    {
     "name": "idle 1",
     "seconds": 5.006072988000028,
     "joules": 364.41,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "reference",
     "seconds": 5.18901718699999,
     "joules": 1149.203,
     "matmuls": 688,
     "dim": 4096,
     "exact": true
    },
    {
     "name": "idle 2",
     "seconds": 5.004339893000065,
     "joules": 367.279,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "control",
     "seconds": 5.00414527800001,
     "joules": 399.381,
     "calls": 2393
    },
    {
     "name": "idle 3",
     "seconds": 5.00391476599998,
     "joules": 365.946,
     "utilization_max": 0,
     "contexts": 1
    },
    {
     "name": "method",
     "seconds": 20.531215340000017,
     "joules": 3749.151,
     "calls": 35,
     "ms": [
      599.5222778320312,
      579.3976440429688,
      579.473388671875,
      579.9321899414062,
      579.8799438476562,
      580.0028076171875,
      580.0284423828125,
      579.9588012695312,
      579.8113403320312,
      579.7908325195312,
      579.789794921875,
      579.7222290039062,
      579.8615112304688,
      579.8154296875,
      579.8379516601562,
      579.9321899414062,
      579.8666381835938,
      579.8983764648438,
      579.80517578125,
      579.8553466796875,
      579.800048828125,
      579.7161254882812,
      579.7990112304688,
      579.72119140625,
      579.7755126953125,
      579.7089233398438,
      579.6710205078125,
      579.5798950195312,
      579.6771850585938,
      579.610595703125,
      579.8379516601562,
      579.6771850585938,
      579.852294921875,
      579.8184814453125,
      579.7805786132812
     ],
     "correct": [
      9803,
      9810,
      9787,
      9798,
      9789,
      9775,
      9792,
      9770,
      9782,
      9766,
      9777,
      9809,
      9794,
      9807,
      9801,
      9774,
      9793,
      9787,
      9773,
      9778,
      9769,
      9798,
      9781,
      9780,
      9805,
      9772,
      9771,
      9787,
      9797,
      9780,
      9807,
      9763,
      9778,
      9801,
      9780
     ]
    },
    {
     "name": "idle 4",
     "seconds": 5.081151716000022,
     "joules": 369.079,
     "utilization_max": 0,
     "contexts": 1
    }
   ]
  }
 }
}
