Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Impl adaptive avgpool #3301

Open
wants to merge 53 commits into
base: develop
Choose a base branch
from
Open

Impl adaptive avgpool #3301

wants to merge 53 commits into from
+3,666 −2

Conversation

hieule88
Copy link
Collaborator

@hieule88 hieule88 commented Oct 7, 2024 •
edited
Loading

  • Added AdaptiveAvgPool 1D 2D 3D forward and backward.

  • Added driver test and gtest for AdaptiveAvgPool .

  • New API is guarded by MIOPEN_BETA_API macro.

  • Average over all cases:

  • AdaptiveAvgPool 1D

Type Forward Backward
float16 2.26 1.94
float32 2.01 1.97
bfloat16 2.25 1.97
  • AdaptiveAvgPool 2D
Type Forward Backward
float16 4.48 1.38
float32 3.15 1.31
bfloat16 3.65 1.39
  • AdaptiveAvgPool 3D
Type Forward Backward
float16 2.14 1.72
float32 2.03 2.06
bfloat16 2.13 1.71
AdaptiveAvgPool 1D
op_name dtype input_size output_size contiguous direction rocm_kernel_avg rocm_op_avg use_miopen MIOPEN MIOPEN_over_Rocm
AdaptiveAvgPool1d float16 [256 256 6] [256 256 6] contiguous fwd 116464 116464 FALSE 10951 10.6350105
AdaptiveAvgPool1d float16 [256 256 6] [256 256 6] contiguous bwd 123199 144335 FALSE 12516 9.84332055
AdaptiveAvgPool1d float16 [256 256 6] [256 256 6] noncontiguous fwd 132927 154559 FALSE 18187 7.308901963
AdaptiveAvgPool1d float16 [256 256 6] [256 256 6] noncontiguous bwd 140752 172063 FALSE 18436 7.634627902
AdaptiveAvgPool1d float16 [16 72 64] [16 72 1] contiguous fwd 12352 12352 FALSE 28035 0.440592117
AdaptiveAvgPool1d float16 [16 72 64] [16 72 1] contiguous bwd 8576 8576 FALSE 4409 1.94511227
AdaptiveAvgPool1d float16 [16 72 64] [16 72 1] noncontiguous fwd 29216 29216 FALSE 33582 0.8699898755
AdaptiveAvgPool1d float16 [16 72 64] [16 72 1] noncontiguous bwd 14336 77104 FALSE 4516 3.1744907
AdaptiveAvgPool1d float16 [16 120 64] [16 120 1] contiguous fwd 12352 12352 FALSE 28106 0.4394791148
AdaptiveAvgPool1d float16 [16 120 64] [16 120 1] contiguous bwd 8448 8448 FALSE 5636 1.498935415
AdaptiveAvgPool1d float16 [16 120 64] [16 120 1] noncontiguous fwd 29104 29104 FALSE 32729 0.8892419567
AdaptiveAvgPool1d float16 [16 120 64] [16 120 1] noncontiguous bwd 16304 72288 FALSE 5547 2.93924644
AdaptiveAvgPool1d float16 [16 480 32] [16 480 1] contiguous fwd 16416 16416 FALSE 12178 1.348004598
AdaptiveAvgPool1d float16 [16 480 32] [16 480 1] contiguous bwd 8352 8352 FALSE 9316 0.8965221125
AdaptiveAvgPool1d float16 [16 480 32] [16 480 1] noncontiguous fwd 21728 21728 FALSE 20853 1.041960389
AdaptiveAvgPool1d float16 [16 480 32] [16 480 1] noncontiguous bwd 20192 69808 FALSE 9991 2.021018917
AdaptiveAvgPool1d float16 [16 672 32] [16 672 1] contiguous fwd 18544 18544 FALSE 12338 1.502998865
AdaptiveAvgPool1d float16 [16 672 32] [16 672 1] contiguous bwd 8464 8464 FALSE 10987 0.7703649768
AdaptiveAvgPool1d float16 [16 672 32] [16 672 1] noncontiguous fwd 20992 20992 FALSE 21226 0.9889757844
AdaptiveAvgPool1d float16 [16 672 32] [16 672 1] noncontiguous bwd 24128 78160 FALSE 11040 2.185507246
AdaptiveAvgPool1d float16 [16 960 32] [16 960 1] contiguous fwd 22560 22560 FALSE 12284 1.836535331
AdaptiveAvgPool1d float16 [16 960 32] [16 960 1] contiguous bwd 8800 8800 FALSE 14667 0.5999863639
AdaptiveAvgPool1d float16 [16 960 32] [16 960 1] noncontiguous fwd 22320 22320 FALSE 22115 1.009269726
AdaptiveAvgPool1d float16 [16 960 32] [16 960 1] noncontiguous bwd 28368 83312 FALSE 14560 1.948351648
AdaptiveAvgPool1d float16 [3 2048 64] [3 2048 1] contiguous fwd 19632 19632 FALSE 27484 0.7143065056
AdaptiveAvgPool1d float16 [3 2048 64] [3 2048 1] contiguous bwd 9040 9040 FALSE 12160 0.7434210526
AdaptiveAvgPool1d float16 [3 2048 64] [3 2048 1] noncontiguous fwd 21424 21424 FALSE 32515 0.6588958942
AdaptiveAvgPool1d float16 [3 2048 64] [3 2048 1] noncontiguous bwd 25376 81008 FALSE 12089 2.099098354
AdaptiveAvgPool1d float16 [64 2208 7] [64 2208 1] contiguous fwd 29200 29200 FALSE 10418 2.802841236
AdaptiveAvgPool1d float16 [64 2208 7] [64 2208 1] contiguous bwd 12384 12384 FALSE 26613 0.4653364897
AdaptiveAvgPool1d float16 [64 2208 7] [64 2208 1] noncontiguous fwd 40240 40240 FALSE 29546 1.361944087
AdaptiveAvgPool1d float16 [64 2208 7] [64 2208 1] noncontiguous bwd 39520 91520 FALSE 26702 1.480038948
AdaptiveAvgPool1d float16 [256 512 14] [256 512 4] contiguous fwd 263138 263138 FALSE 17227 15.27474314
AdaptiveAvgPool1d float16 [256 512 14] [256 512 4] contiguous bwd 336223 352543 FALSE 60995 5.512304287
AdaptiveAvgPool1d float16 [256 512 14] [256 512 4] noncontiguous fwd 304639 321151 FALSE 58062 5.246787916
AdaptiveAvgPool1d float16 [256 512 14] [256 512 4] noncontiguous bwd 392126 426430 FALSE 61777 6.347443223
AdaptiveAvgPool1d float16 [256 528 14] [256 528 4] contiguous fwd 260223 260223 FALSE 17298 15.04353104
AdaptiveAvgPool1d float16 [256 528 14] [256 528 4] contiguous bwd 347679 365662 FALSE 63040 5.515212563
AdaptiveAvgPool1d float16 [256 528 14] [256 528 4] noncontiguous fwd 313903 330463 FALSE 60782 5.164407226
AdaptiveAvgPool1d float16 [256 528 14] [256 528 4] noncontiguous bwd 401582 435118 FALSE 63714 6.302884766
AdaptiveAvgPool1d float16 [256 1024 7] [256 1024 1] contiguous fwd 44336 44336 FALSE 17618 2.516517198
AdaptiveAvgPool1d float16 [256 1024 7] [256 1024 1] contiguous bwd 16480 16480 FALSE 46471 0.3546297691
AdaptiveAvgPool1d float16 [256 1024 7] [256 1024 1] noncontiguous fwd 68320 68320 FALSE 55413 1.232923682
AdaptiveAvgPool1d float16 [256 1024 7] [256 1024 1] noncontiguous bwd 71103 113631 FALSE 46649 1.524212738
AdaptiveAvgPool1d float16 [64 768 1] [64 768 1] contiguous fwd 10928 10928 FALSE 6187 1.766284144
AdaptiveAvgPool1d float16 [64 768 1] [64 768 1] contiguous bwd 5344 5344 FALSE 6258 0.8539469479
AdaptiveAvgPool1d float16 [64 768 1] [64 768 1] noncontiguous fwd 11920 11920 FALSE 7022 1.697522073
AdaptiveAvgPool1d float16 [64 768 1] [64 768 1] noncontiguous bwd 13584 55776 FALSE 7058 1.92462454
AdaptiveAvgPool1d float16 [64 2048 8] [64 2048 1] contiguous fwd 35904 35904 FALSE 13724 2.616146896
AdaptiveAvgPool1d float16 [64 2048 8] [64 2048 1] contiguous bwd 12736 12736 FALSE 27840 0.4574712644
AdaptiveAvgPool1d float16 [64 2048 8] [64 2048 1] noncontiguous fwd 38272 38272 FALSE 29457 1.299249754
AdaptiveAvgPool1d float16 [64 2048 8] [64 2048 1] noncontiguous bwd 45808 105999 FALSE 27769 1.649609277
AdaptiveAvgPool1d float16 [64 72 22] [64 72 1] contiguous fwd 12640 12640 FALSE 10027 1.26059639
AdaptiveAvgPool1d float16 [64 72 22] [64 72 1] contiguous bwd 8160 8160 FALSE 5973 1.366147664
AdaptiveAvgPool1d float16 [64 72 22] [64 72 1] noncontiguous fwd 24800 24800 FALSE 16800 1.476190476
AdaptiveAvgPool1d float16 [64 72 22] [64 72 1] noncontiguous bwd 17184 83232 FALSE 6293 2.730653107
AdaptiveAvgPool1d float16 [64 120 22] [64 120 1] contiguous fwd 13056 13056 FALSE 9724 1.342657343
AdaptiveAvgPool1d float16 [64 120 22] [64 120 1] contiguous bwd 8208 8208 FALSE 7111 1.154268035
AdaptiveAvgPool1d float16 [64 120 22] [64 120 1] noncontiguous fwd 25600 25600 FALSE 17600 1.454545455
AdaptiveAvgPool1d float16 [64 120 22] [64 120 1] noncontiguous bwd 17664 59872 FALSE 6987 2.528123658
AdaptiveAvgPool1d float16 [64 480 11] [64 480 1] contiguous fwd 19968 19968 FALSE 8284 2.410429744
AdaptiveAvgPool1d float16 [64 480 11] [64 480 1] contiguous bwd 8448 8448 FALSE 11147 0.7578720732
AdaptiveAvgPool1d float16 [64 480 11] [64 480 1] noncontiguous fwd 21344 21344 FALSE 15822 1.349007711
AdaptiveAvgPool1d float16 [64 480 11] [64 480 1] noncontiguous bwd 21216 61424 FALSE 11484 1.847439916
AdaptiveAvgPool1d float16 [64 672 11] [64 672 1] contiguous fwd 22544 22544 FALSE 8587 2.625363922
AdaptiveAvgPool1d float16 [64 672 11] [64 672 1] contiguous bwd 8672 8672 FALSE 14204 0.6105322444
AdaptiveAvgPool1d float16 [64 672 11] [64 672 1] noncontiguous fwd 23504 23504 FALSE 17920 1.311607143
AdaptiveAvgPool1d float16 [64 672 11] [64 672 1] noncontiguous bwd 24032 70608 FALSE 14311 1.679267696
AdaptiveAvgPool1d float16 [64 960 11] [64 960 1] contiguous fwd 27808 27808 FALSE 9920 2.803225806
AdaptiveAvgPool1d float16 [64 960 11] [64 960 1] contiguous bwd 9264 9264 FALSE 19058 0.4860950782
AdaptiveAvgPool1d float16 [64 960 11] [64 960 1] noncontiguous fwd 33296 33296 FALSE 23627 1.409235197
AdaptiveAvgPool1d float16 [64 960 11] [64 960 1] noncontiguous bwd 31328 74160 FALSE 19075 1.642359109
AdaptiveAvgPool1d float16 [55 72 14] [55 72 1] contiguous fwd 11680 11680 FALSE 8871 1.316649758
AdaptiveAvgPool1d float16 [55 72 14] [55 72 1] contiguous bwd 8000 8000 FALSE 5849 1.367755172
AdaptiveAvgPool1d float16 [55 72 14] [55 72 1] noncontiguous fwd 16480 16480 FALSE 14560 1.131868132
AdaptiveAvgPool1d float16 [55 72 14] [55 72 1] noncontiguous bwd 15264 69536 FALSE 5956 2.562793821
AdaptiveAvgPool1d float16 [55 120 14] [55 120 1] contiguous fwd 12000 12000 FALSE 8604 1.394700139
AdaptiveAvgPool1d float16 [55 120 14] [55 120 1] contiguous bwd 8144 8144 FALSE 5902 1.37987123
AdaptiveAvgPool1d float16 [55 120 14] [55 120 1] noncontiguous fwd 19200 19200 FALSE 14667 1.309061158
AdaptiveAvgPool1d float16 [55 120 14] [55 120 1] noncontiguous bwd 15808 54032 FALSE 6400 2.47
AdaptiveAvgPool1d float16 [55 480 7] [55 480 1] contiguous fwd 12880 12880 FALSE 7307 1.762693308
AdaptiveAvgPool1d float16 [55 480 7] [55 480 1] contiguous bwd 8608 8608 FALSE 7271 1.183881172
AdaptiveAvgPool1d float16 [55 480 7] [55 480 1] noncontiguous fwd 16128 16128 FALSE 10915 1.477599634
AdaptiveAvgPool1d float16 [55 480 7] [55 480 1] noncontiguous bwd 17840 56368 FALSE 7164 2.490228922
AdaptiveAvgPool1d float16 [55 672 7] [55 672 1] contiguous fwd 13856 13856 FALSE 7413 1.869148793
AdaptiveAvgPool1d float16 [55 672 7] [55 672 1] contiguous bwd 8656 8656 FALSE 10240 0.8453125
AdaptiveAvgPool1d float16 [55 672 7] [55 672 1] noncontiguous fwd 17520 17520 FALSE 12320 1.422077922
AdaptiveAvgPool1d float16 [55 672 7] [55 672 1] noncontiguous bwd 19152 62032 FALSE 10382 1.844731266
AdaptiveAvgPool1d float16 [55 960 7] [55 960 1] contiguous fwd 16112 16112 FALSE 7662 2.10284521
AdaptiveAvgPool1d float16 [55 960 7] [55 960 1] contiguous bwd 8768 8768 FALSE 11964 0.7328652625
AdaptiveAvgPool1d float16 [55 960 7] [55 960 1] noncontiguous fwd 19984 19984 FALSE 14738 1.355950604
AdaptiveAvgPool1d float16 [55 960 7] [55 960 1] noncontiguous bwd 21408 84832 FALSE 11787 1.816238229
AdaptiveAvgPool1d float16 [64 72 19] [64 72 1] contiguous fwd 12480 12480 FALSE 9742 1.281051119
AdaptiveAvgPool1d float16 [64 72 19] [64 72 1] contiguous bwd 8000 8000 FALSE 5920 1.351351351
AdaptiveAvgPool1d float16 [64 72 19] [64 72 1] noncontiguous fwd 23424 23424 FALSE 15715 1.49055043
AdaptiveAvgPool1d float16 [64 72 19] [64 72 1] noncontiguous bwd 17056 86511 FALSE 6258 2.725471397
AdaptiveAvgPool1d float16 [64 120 19] [64 120 1] contiguous fwd 13232 13232 FALSE 9582 1.380922563
AdaptiveAvgPool1d float16 [64 120 19] [64 120 1] contiguous bwd 8032 8032 FALSE 6631 1.21128035
AdaptiveAvgPool1d float16 [64 120 19] [64 120 1] noncontiguous fwd 24448 24448 FALSE 15963 1.53154169
AdaptiveAvgPool1d float16 [64 120 19] [64 120 1] noncontiguous bwd 17568 73104 FALSE 6471 2.71488178
AdaptiveAvgPool1d float16 [64 480 10] [64 480 1] contiguous fwd 19984 19984 FALSE 7822 2.554845308
AdaptiveAvgPool1d float16 [64 480 10] [64 480 1] contiguous bwd 8656 8656 FALSE 10809 0.8008141364
AdaptiveAvgPool1d float16 [64 480 10] [64 480 1] noncontiguous fwd 20176 20176 FALSE 14951 1.349474952
AdaptiveAvgPool1d float16 [64 480 10] [64 480 1] noncontiguous bwd 21632 73808 FALSE 11218 1.92832947
AdaptiveAvgPool1d float16 [64 672 10] [64 672 1] contiguous fwd 22480 22480 FALSE 8391 2.679060899
AdaptiveAvgPool1d float16 [64 672 10] [64 672 1] contiguous bwd 8704 8704 FALSE 13315 0.6536988359
AdaptiveAvgPool1d float16 [64 672 10] [64 672 1] noncontiguous fwd 22112 22112 FALSE 16835 1.313454113
AdaptiveAvgPool1d float16 [64 672 10] [64 672 1] noncontiguous bwd 24544 80896 FALSE 13440 1.826190476
AdaptiveAvgPool1d float16 [64 960 10] [64 960 1] contiguous fwd 27440 27440 FALSE 9049 3.032379268
AdaptiveAvgPool1d float16 [64 960 10] [64 960 1] contiguous bwd 9200 9200 FALSE 17600 0.5227272727
AdaptiveAvgPool1d float16 [64 960 10] [64 960 1] noncontiguous fwd 31040 31040 FALSE 21991 1.411486517
AdaptiveAvgPool1d float16 [64 960 10] [64 960 1] noncontiguous bwd 32592 101312 FALSE 17564 1.855613755
AdaptiveAvgPool1d float16 [64 72 25] [64 72 1] contiguous fwd 12624 12624 FALSE 10738 1.175637921
AdaptiveAvgPool1d float16 [64 72 25] [64 72 1] contiguous bwd 8048 8048 FALSE 6027 1.335324374
AdaptiveAvgPool1d float16 [64 72 25] [64 72 1] noncontiguous fwd 26912 26912 FALSE 18382 1.46404091
AdaptiveAvgPool1d float16 [64 72 25] [64 72 1] noncontiguous bwd 17200 69760 FALSE 6400 2.6875
AdaptiveAvgPool1d float16 [64 120 25] [64 120 1] contiguous fwd 13424 13424 FALSE 10293 1.304187312
AdaptiveAvgPool1d float16 [64 120 25] [64 120 1] contiguous bwd 8224 8224 FALSE 7787 1.056119173
AdaptiveAvgPool1d float16 [64 120 25] [64 120 1] noncontiguous fwd 28016 28016 FALSE 18702 1.498021602
AdaptiveAvgPool1d float16 [64 120 25] [64 120 1] noncontiguous bwd 17936 91776 FALSE 7662 2.340903158
AdaptiveAvgPool1d float16 [64 480 13] [64 480 1] contiguous fwd 19968 19968 FALSE 8782 2.273741744
AdaptiveAvgPool1d float16 [64 480 13] [64 480 1] contiguous bwd 8592 8592 FALSE 12444 0.6904532305
AdaptiveAvgPool1d float16 [64 480 13] [64 480 1] noncontiguous fwd 22192 22192 FALSE 17671 1.255842906
AdaptiveAvgPool1d float16 [64 480 13] [64 480 1] noncontiguous bwd 23296 79392 FALSE 12373 1.882809343
AdaptiveAvgPool1d float16 [64 672 13] [64 672 1] contiguous fwd 22672 22672 FALSE 8924 2.540564769
AdaptiveAvgPool1d float16 [64 672 13] [64 672 1] contiguous bwd 8848 8848 FALSE 16230 0.5451632779
AdaptiveAvgPool1d float16 [64 672 13] [64 672 1] noncontiguous fwd 25168 25168 FALSE 20213 1.245139267
AdaptiveAvgPool1d float16 [64 672 13] [64 672 1] noncontiguous bwd 27152 85920 FALSE 16000 1.697
AdaptiveAvgPool1d float16 [64 960 13] [64 960 1] contiguous fwd 27760 27760 FALSE 11164 2.486563956
AdaptiveAvgPool1d float16 [64 960 13] [64 960 1] contiguous bwd 9552 9552 FALSE 21742 0.4393340079
AdaptiveAvgPool1d float16 [64 960 13] [64 960 1] noncontiguous fwd 36144 36144 FALSE 26755 1.350925061
AdaptiveAvgPool1d float16 [64 960 13] [64 960 1] noncontiguous bwd 35408 88096 FALSE 21778 1.625860961
AdaptiveAvgPool1d float16 [54 72 25] [54 72 1] contiguous fwd 11824 11824 FALSE 10613 1.114105343
AdaptiveAvgPool1d float16 [54 72 25] [54 72 1] contiguous bwd 8448 8448 FALSE 5831 1.448808095
AdaptiveAvgPool1d float16 [54 72 25] [54 72 1] noncontiguous fwd 20768 20768 FALSE 19804 1.048677035
AdaptiveAvgPool1d float16 [54 72 25] [54 72 1] noncontiguous bwd 15840 62768 FALSE 6169 2.567677095
AdaptiveAvgPool1d float16 [47 72 25] [47 72 1] contiguous fwd 12064 12064 FALSE 11093 1.087532678
AdaptiveAvgPool1d float16 [47 72 25] [47 72 1] contiguous bwd 8256 8256 FALSE 5511 1.49809472
AdaptiveAvgPool1d float16 [47 72 25] [47 72 1] noncontiguous fwd 20352 20352 FALSE 20053 1.014910487
AdaptiveAvgPool1d float16 [47 72 25] [47 72 1] noncontiguous bwd 15584 64320 FALSE 5867 2.656212715
AdaptiveAvgPool1d float16 [54 120 25] [54 120 1] contiguous fwd 12112 12112 FALSE 10222 1.184895324
AdaptiveAvgPool1d float16 [54 120 25] [54 120 1] contiguous bwd 8208 8208 FALSE 7129 1.151353626
AdaptiveAvgPool1d float16 [54 120 25] [54 120 1] noncontiguous fwd 22784 22784 FALSE 19502 1.168290432
AdaptiveAvgPool1d float16 [54 120 25] [54 120 1] noncontiguous bwd 17328 67072 FALSE 6898 2.512032473
AdaptiveAvgPool1d float16 [47 120 25] [47 120 1] contiguous fwd 12352 12352 FALSE 10471 1.179639003
AdaptiveAvgPool1d float16 [47 120 25] [47 120 1] contiguous bwd 8176 8176 FALSE 6364 1.284726587
AdaptiveAvgPool1d float16 [47 120 25] [47 120 1] noncontiguous fwd 23136 23136 FALSE 19769 1.170317163
AdaptiveAvgPool1d float16 [47 120 25] [47 120 1] noncontiguous bwd 17120 53008 FALSE 6471 2.645649822
AdaptiveAvgPool1d float16 [54 480 13] [54 480 1] contiguous fwd 16416 16416 FALSE 8284 1.981651376
AdaptiveAvgPool1d float16 [54 480 13] [54 480 1] contiguous bwd 8304 8304 FALSE 11111 0.7473674737
AdaptiveAvgPool1d float16 [54 480 13] [54 480 1] noncontiguous fwd 19760 19760 FALSE 15253 1.295482856
AdaptiveAvgPool1d float16 [54 480 13] [54 480 1] noncontiguous bwd 20800 61024 FALSE 11449 1.816752555
AdaptiveAvgPool1d float16 [47 480 13] [47 480 1] contiguous fwd 15104 15104 FALSE 8498 1.777359379
AdaptiveAvgPool1d float16 [47 480 13] [47 480 1] contiguous bwd 8272 8272 FALSE 10453 0.791351765
AdaptiveAvgPool1d float16 [47 480 13] [47 480 1] noncontiguous fwd 19552 19552 FALSE 15218 1.284794323
AdaptiveAvgPool1d float16 [47 480 13] [47 480 1] noncontiguous bwd 19072 50736 FALSE 10880 1.752941176
AdaptiveAvgPool1d float16 [54 672 13] [54 672 1] contiguous fwd 21312 21312 FALSE 8693 2.451627746
AdaptiveAvgPool1d float16 [54 672 13] [54 672 1] contiguous bwd 9008 9008 FALSE 14257 0.6318299783
AdaptiveAvgPool1d float16 [54 672 13] [54 672 1] noncontiguous fwd 22464 22464 FALSE 18275 1.229220246
AdaptiveAvgPool1d float16 [54 672 13] [54 672 1] noncontiguous bwd 24080 80815 FALSE 14115 1.705986539
AdaptiveAvgPool1d float16 [47 672 13] [47 672 1] contiguous fwd 19952 19952 FALSE 8533 2.338216337
AdaptiveAvgPool1d float16 [47 672 13] [47 672 1] contiguous bwd 8448 8448 FALSE 12942 0.6527584608
AdaptiveAvgPool1d float16 [47 672 13] [47 672 1] noncontiguous fwd 26784 26784 FALSE 17422 1.537366548
AdaptiveAvgPool1d float16 [47 672 13] [47 672 1] noncontiguous bwd 23920 63072 FALSE 12693 1.88450327
AdaptiveAvgPool1d float16 [47 960 13] [47 960 1] contiguous fwd 23008 23008 FALSE 8924 2.578216047
AdaptiveAvgPool1d float16 [47 960 13] [47 960 1] contiguous bwd 8880 8880 FALSE 16853 0.5269091556
AdaptiveAvgPool1d float16 [47 960 13] [47 960 1] noncontiguous fwd 25152 25152 FALSE 20391 1.233485361
AdaptiveAvgPool1d float16 [47 960 13] [47 960 1] noncontiguous bwd 26704 74928 FALSE 16711 1.597989348
AdaptiveAvgPool1d float16 [54 960 13] [54 960 1] contiguous fwd 24688 24688 FALSE 9102 2.712370907
AdaptiveAvgPool1d float16 [54 960 13] [54 960 1] contiguous bwd 9232 9232 FALSE 19022 0.4853327726
AdaptiveAvgPool1d float16 [54 960 13] [54 960 1] noncontiguous fwd 26976 26976 FALSE 22329 1.208115007
AdaptiveAvgPool1d float16 [54 960 13] [54 960 1] noncontiguous bwd 29968 83216 FALSE 19075 1.571061599
AdaptiveAvgPool1d float16 [512 1280 7] [512 1280 1] contiguous fwd 91888 91888 FALSE 26862 3.420743057
AdaptiveAvgPool1d float16 [512 1280 7] [512 1280 1] contiguous bwd 30912 30912 FALSE 112178 0.2755620532
AdaptiveAvgPool1d float16 [512 1280 7] [512 1280 1] noncontiguous fwd 157343 157343 FALSE 126151 1.247259237
AdaptiveAvgPool1d float16 [512 1280 7] [512 1280 1] noncontiguous bwd 170895 204767 FALSE 112764 1.515510269
AdaptiveAvgPool1d float16 [256 1280 7] [256 1280 1] contiguous fwd 53440 53440 FALSE 18720 2.854700855
AdaptiveAvgPool1d float16 [256 1280 7] [256 1280 1] contiguous bwd 18624 18624 FALSE 57991 0.3211532824
AdaptiveAvgPool1d float16 [256 1280 7] [256 1280 1] noncontiguous fwd 86144 86144 FALSE 64746 1.330491459
AdaptiveAvgPool1d float16 [256 1280 7] [256 1280 1] noncontiguous bwd 85856 128064 FALSE 57938 1.481859919
AdaptiveAvgPool1d float16 [256 16 56] [256 16 1] contiguous fwd 13552 13552 FALSE 18471 0.7336906502
AdaptiveAvgPool1d float16 [256 16 56] [256 16 1] contiguous bwd 8032 8032 FALSE 8356 0.9612254667
AdaptiveAvgPool1d float16 [256 16 56] [256 16 1] noncontiguous fwd 34880 34880 FALSE 32569 1.070957045
AdaptiveAvgPool1d float16 [256 16 56] [256 16 1] noncontiguous bwd 20896 69680 FALSE 8907 2.346019984
AdaptiveAvgPool1d float16 [256 96 14] [256 96 1] contiguous fwd 16144 16144 FALSE 8711 1.853288945
AdaptiveAvgPool1d float16 [256 96 14] [256 96 1] contiguous bwd 8544 8544 FALSE 11218 0.7616330897
AdaptiveAvgPool1d float16 [256 96 14] [256 96 1] noncontiguous fwd 25968 25968 FALSE 15591 1.665576294
AdaptiveAvgPool1d float16 [256 96 14] [256 96 1] noncontiguous bwd 23744 69648 FALSE 11502 2.064336637
AdaptiveAvgPool1d float16 [256 240 14] [256 240 1] contiguous fwd 27792 27792 FALSE 11947 2.326274379
AdaptiveAvgPool1d float16 [256 240 14] [256 240 1] contiguous bwd 11184 11184 FALSE 23484 0.4762391415
AdaptiveAvgPool1d float16 [256 240 14] [256 240 1] noncontiguous fwd 44528 44528 FALSE 30578 1.456210347
AdaptiveAvgPool1d float16 [256 240 14] [256 240 1] noncontiguous bwd 39888 86080 FALSE 23449 1.70105335
AdaptiveAvgPool1d float16 [256 120 14] [256 120 1] contiguous fwd 20112 20112 FALSE 8942 2.249161261
AdaptiveAvgPool1d float16 [256 120 14] [256 120 1] contiguous bwd 8752 8752 FALSE 13404 0.6529394211
AdaptiveAvgPool1d float16 [256 120 14] [256 120 1] noncontiguous fwd 26832 26832 FALSE 22133 1.212307414
AdaptiveAvgPool1d float16 [256 120 14] [256 120 1] noncontiguous bwd 26368 70976 FALSE 13298 1.982854565
AdaptiveAvgPool1d float16 [256 144 14] [256 144 1] contiguous fwd 21504 21504 FALSE 9156 2.348623853
AdaptiveAvgPool1d float16 [256 144 14] [256 144 1] contiguous bwd 9008 9008 FALSE 15484 0.5817618187
AdaptiveAvgPool1d float16 [256 144 14] [256 144 1] noncontiguous fwd 27568 27568 FALSE 22044 1.25058973
AdaptiveAvgPool1d float16 [256 144 14] [256 144 1] noncontiguous bwd 28688 70336 FALSE 15236 1.8829089
AdaptiveAvgPool1d float16 [256 288 7] [256 288 1] contiguous fwd 21680 21680 FALSE 8160 2.656862745
AdaptiveAvgPool1d float16 [256 288 7] [256 288 1] contiguous bwd 9360 9360 FALSE 15324 0.6108065779
AdaptiveAvgPool1d float16 [256 288 7] [256 288 1] noncontiguous fwd 30016 30016 FALSE 19751 1.51972052
AdaptiveAvgPool1d float16 [256 288 7] [256 288 1] noncontiguous bwd 29360 70240 FALSE 15467 1.898234952
AdaptiveAvgPool1d float16 [256 576 7] [256 576 1] contiguous fwd 30256 30256 FALSE 11004 2.74954562
AdaptiveAvgPool1d float16 [256 576 7] [256 576 1] contiguous bwd 12816 12816 FALSE 27573 0.4648025242
AdaptiveAvgPool1d float16 [256 576 7] [256 576 1] noncontiguous fwd 43120 43120 FALSE 33084 1.303349051
AdaptiveAvgPool1d float16 [256 576 7] [256 576 1] noncontiguous bwd 45824 87152 FALSE 27555 1.66300127
AdaptiveAvgPool1d float16 [256 2048 7] [256 2048 1] contiguous fwd 76207 76207 FALSE 23093 3.30000433
AdaptiveAvgPool1d float16 [256 2048 7] [256 2048 1] contiguous bwd 25584 25584 FALSE 90080 0.2840142096
AdaptiveAvgPool1d float16 [256 2048 7] [256 2048 1] noncontiguous fwd 128319 128319 FALSE 97457 1.316672994
AdaptiveAvgPool1d float16 [256 2048 7] [256 2048 1] noncontiguous bwd 143936 181151 FALSE 90595 1.588785253
AdaptiveAvgPool1d float16 [1536 100 13] [1536 100 1] contiguous fwd 51008 51008 FALSE 24249 2.103509423
AdaptiveAvgPool1d float16 [1536 100 13] [1536 100 1] contiguous bwd 16992 16992 FALSE 50275 0.3379811039
AdaptiveAvgPool1d float16 [1536 100 13] [1536 100 1] noncontiguous fwd 66736 66736 FALSE 53191 1.254648343
AdaptiveAvgPool1d float16 [1536 100 13] [1536 100 1] noncontiguous bwd 73632 123807 FALSE 50524 1.457366796
AdaptiveAvgPool1d float16 [731 100 13] [731 100 1] contiguous fwd 30784 30784 FALSE 13618 2.260537524
AdaptiveAvgPool1d float16 [731 100 13] [731 100 1] contiguous bwd 11744 11744 FALSE 25475 0.4610009814
AdaptiveAvgPool1d float16 [731 100 13] [731 100 1] noncontiguous fwd 29152 29152 FALSE 16515 1.765183167
AdaptiveAvgPool1d float16 [731 100 13] [731 100 1] noncontiguous bwd 43152 91423 FALSE 25422 1.697427425
AdaptiveAvgPool1d float16 [724 100 13] [724 100 1] contiguous fwd 30544 30544 FALSE 13262 2.303121701
AdaptiveAvgPool1d float16 [724 100 13] [724 100 1] contiguous bwd 11840 11840 FALSE 25351 0.4670427202
AdaptiveAvgPool1d float16 [724 100 13] [724 100 1] noncontiguous fwd 28960 28960 FALSE 16409 1.764885124
AdaptiveAvgPool1d float16 [724 100 13] [724 100 1] noncontiguous bwd 42384 88527 FALSE 25475 1.663748773
AdaptiveAvgPool1d float16 [64 16 80] [64 16 1] contiguous fwd 13136 13136 FALSE 23964 0.5481555667
AdaptiveAvgPool1d float16 [64 16 80] [64 16 1] contiguous bwd 8656 8656 FALSE 4693 1.84444918
AdaptiveAvgPool1d float16 [64 16 80] [64 16 1] noncontiguous fwd 36368 36368 FALSE 42453 0.8566650178
AdaptiveAvgPool1d float16 [64 16 80] [64 16 1] noncontiguous bwd 16352 64624 FALSE 4622 3.537862397
AdaptiveAvgPool1d float16 [64 96 20] [64 96 1] contiguous fwd 13296 13296 FALSE 9724 1.367338544
AdaptiveAvgPool1d float16 [64 96 20] [64 96 1] contiguous bwd 8608 8608 FALSE 6187 1.391304348
AdaptiveAvgPool1d float16 [64 96 20] [64 96 1] noncontiguous fwd 23920 23920 FALSE 15786 1.515266692
AdaptiveAvgPool1d float16 [64 96 20] [64 96 1] noncontiguous bwd 17920 61056 FALSE 6436 2.784338098
AdaptiveAvgPool1d float16 [64 240 20] [64 240 1] contiguous fwd 20064 20064 FALSE 9724 2.063348416
AdaptiveAvgPool1d float16 [64 240 20] [64 240 1] contiguous bwd 8560 8560 FALSE 10524 0.8133789434
AdaptiveAvgPool1d float16 [64 240 20] [64 240 1] noncontiguous fwd 24976 24976 FALSE 17280 1.44537037
AdaptiveAvgPool1d float16 [64 240 20] [64 240 1] noncontiguous bwd 21808 84112 FALSE 10844 2.011066027
AdaptiveAvgPool1d float16 [64 120 20] [64 120 1] contiguous fwd 13984 13984 FALSE 9707 1.440609869
AdaptiveAvgPool1d float16 [64 120 20] [64 120 1] contiguous bwd 8608 8608 FALSE 6471 1.330242621
AdaptiveAvgPool1d float16 [64 120 20] [64 120 1] noncontiguous fwd 24256 24256 FALSE 16587 1.462350033
AdaptiveAvgPool1d float16 [64 120 20] [64 120 1] noncontiguous bwd 17936 65632 FALSE 6596 2.719223772
AdaptiveAvgPool1d float16 [64 144 20] [64 144 1] contiguous fwd 14448 14448 FALSE 9582 1.507827176
AdaptiveAvgPool1d float16 [64 144 20] [64 144 1] contiguous bwd 8448 8448 FALSE 7182 1.176274018
AdaptiveAvgPool1d float16 [64 144 20] [64 144 1] noncontiguous fwd 24288 24288 FALSE 16604 1.462780053
AdaptiveAvgPool1d float16 [64 144 20] [64 144 1] noncontiguous bwd 18800 64672 FALSE 7076 2.656868287
AdaptiveAvgPool1d float16 [64 288 10] [64 288 1] contiguous fwd 14304 14304 FALSE 7911 1.808115283
AdaptiveAvgPool1d float16 [64 288 10] [64 288 1] contiguous bwd 8720 8720 FALSE 7253 1.202261133
AdaptiveAvgPool1d float16 [64 288 10] [64 288 1] noncontiguous fwd 19840 19840 FALSE 12142 1.633997694
AdaptiveAvgPool1d float16 [64 288 10] [64 288 1] noncontiguous bwd 18480 66384 FALSE 7200 2.566666667
AdaptiveAvgPool1d float16 [256 512 7] [256 512 7] contiguous fwd 227231 227231 FALSE 20018 11.3513338
AdaptiveAvgPool1d float16 [256 512 7] [256 512 7] contiguous bwd 238671 254527 FALSE 24960 9.562139423
AdaptiveAvgPool1d float16 [256 512 7] [256 512 7] noncontiguous fwd 257983 274367 FALSE 31129 8.287545376
AdaptiveAvgPool1d float16 [256 512 7] [256 512 7] noncontiguous bwd 270079 302446 FALSE 31626 8.539777398
AdaptiveAvgPool1d float32 [256 256 6] [256 256 6] contiguous fwd 115759 115759 FALSE 11342 10.20622465
AdaptiveAvgPool1d float32 [256 256 6] [256 256 6] contiguous bwd 111247 128015 FALSE 12640 8.801186709
AdaptiveAvgPool1d float32 [256 256 6] [256 256 6] noncontiguous fwd 133919 150303 FALSE 18702 7.160678002
AdaptiveAvgPool1d float32 [256 256 6] [256 256 6] noncontiguous bwd 129567 161631 FALSE 19004 6.817880446
AdaptiveAvgPool1d float32 [16 72 64] [16 72 1] contiguous fwd 11728 11728 FALSE 38222 0.3068389932
AdaptiveAvgPool1d float32 [16 72 64] [16 72 1] contiguous bwd 6816 6816 FALSE 5013 1.359664871
AdaptiveAvgPool1d float32 [16 72 64] [16 72 1] noncontiguous fwd 37232 37232 FALSE 34489 1.079532605
AdaptiveAvgPool1d float32 [16 72 64] [16 72 1] noncontiguous bwd 16192 68096 FALSE 5156 3.140418929
AdaptiveAvgPool1d float32 [16 120 64] [16 120 1] contiguous fwd 12704 12704 FALSE 37529 0.3385115511
AdaptiveAvgPool1d float32 [16 120 64] [16 120 1] contiguous bwd 7488 7488 FALSE 5813 1.288147256
AdaptiveAvgPool1d float32 [16 120 64] [16 120 1] noncontiguous fwd 41360 41360 FALSE 37155 1.113174539
AdaptiveAvgPool1d float32 [16 120 64] [16 120 1] noncontiguous bwd 18608 66272 FALSE 6098 3.051492293
AdaptiveAvgPool1d float32 [16 480 32] [16 480 1] contiguous fwd 16912 16912 FALSE 17440 0.9697247706
AdaptiveAvgPool1d float32 [16 480 32] [16 480 1] contiguous bwd 8128 8128 FALSE 9067 0.896437631
AdaptiveAvgPool1d float32 [16 480 32] [16 480 1] noncontiguous fwd 31328 31328 FALSE 22204 1.410916952
AdaptiveAvgPool1d float32 [16 480 32] [16 480 1] noncontiguous bwd 21440 66128 FALSE 9956 2.153475291
AdaptiveAvgPool1d float32 [16 672 32] [16 672 1] contiguous fwd 18976 18976 FALSE 17760 1.068468468
AdaptiveAvgPool1d float32 [16 672 32] [16 672 1] contiguous bwd 8512 8512 FALSE 11075 0.7685778781
AdaptiveAvgPool1d float32 [16 672 32] [16 672 1] noncontiguous fwd 31584 31584 FALSE 22702 1.391243062
AdaptiveAvgPool1d float32 [16 672 32] [16 672 1] noncontiguous bwd 25472 73504 FALSE 11200 2.274285714
AdaptiveAvgPool1d float32 [16 960 32] [16 960 1] contiguous fwd 22768 22768 FALSE 20498 1.110742511
AdaptiveAvgPool1d float32 [16 960 32] [16 960 1] contiguous bwd 9056 9056 FALSE 14649 0.6181991945
AdaptiveAvgPool1d float32 [16 960 32] [16 960 1] noncontiguous fwd 31776 31776 FALSE 23858 1.331880292
AdaptiveAvgPool1d float32 [16 960 32] [16 960 1] noncontiguous bwd 29504 80338 FALSE 14471 2.038836293
AdaptiveAvgPool1d float32 [3 2048 64] [3 2048 1] contiguous fwd 19904 19904 FALSE 38417 0.5181039644
AdaptiveAvgPool1d float32 [3 2048 64] [3 2048 1] contiguous bwd 8560 8560 FALSE 12124 0.7060376113
AdaptiveAvgPool1d float32 [3 2048 64] [3 2048 1] noncontiguous fwd 20416 20416 FALSE 33280 0.6134615385
AdaptiveAvgPool1d float32 [3 2048 64] [3 2048 1] noncontiguous bwd 26768 68704 FALSE 12000 2.230666667
AdaptiveAvgPool1d float32 [64 2208 7] [64 2208 1] contiguous fwd 34016 34016 FALSE 17209 1.97664013
AdaptiveAvgPool1d float32 [64 2208 7] [64 2208 1] contiguous bwd 12816 12816 FALSE 26489 0.4838234739
AdaptiveAvgPool1d float32 [64 2208 7] [64 2208 1] noncontiguous fwd 47536 47536 FALSE 32550 1.460399386
AdaptiveAvgPool1d float32 [64 2208 7] [64 2208 1] noncontiguous bwd 45488 90912 FALSE 26649 1.706930842
AdaptiveAvgPool1d float32 [256 512 14] [256 512 4] contiguous fwd 249999 249999 FALSE 18755 13.32972541
AdaptiveAvgPool1d float32 [256 512 14] [256 512 4] contiguous bwd 230815 247887 FALSE 60853 3.79299295
AdaptiveAvgPool1d float32 [256 512 14] [256 512 4] noncontiguous fwd 306927 323567 FALSE 61528 4.98841178
AdaptiveAvgPool1d float32 [256 512 14] [256 512 4] noncontiguous bwd 308735 343391 FALSE 61386 5.029404099
AdaptiveAvgPool1d float32 [256 528 14] [256 528 4] contiguous fwd 258959 258959 FALSE 19147 13.52478195
AdaptiveAvgPool1d float32 [256 528 14] [256 528 4] contiguous bwd 239199 256575 FALSE 62737 3.812726142
AdaptiveAvgPool1d float32 [256 528 14] [256 528 4] noncontiguous fwd 316398 332510 FALSE 63590 4.975593647
AdaptiveAvgPool1d float32 [256 528 14] [256 528 4] noncontiguous bwd 308398 342222 FALSE 63519 4.855208678
AdaptiveAvgPool1d float32 [256 1024 7] [256 1024 1] contiguous fwd 53072 53072 FALSE 28942 1.833736438
AdaptiveAvgPool1d float32 [256 1024 7] [256 1024 1] contiguous bwd 18144 18144 FALSE 46577 0.3895484896
AdaptiveAvgPool1d float32 [256 1024 7] [256 1024 1] noncontiguous fwd 73232 73232 FALSE 58897 1.243391005
AdaptiveAvgPool1d float32 [256 1024 7] [256 1024 1] noncontiguous bwd 94992 135663 FALSE 46755 2.031697145
AdaptiveAvgPool1d float32 [64 768 1] [64 768 1] contiguous fwd 12608 12608 FALSE 6151 2.049748008
AdaptiveAvgPool1d float32 [64 768 1] [64 768 1] contiguous bwd 4960 4960 FALSE 6329 0.7836941065
AdaptiveAvgPool1d float32 [64 768 1] [64 768 1] noncontiguous fwd 12576 12576 FALSE 9244 1.360450022
AdaptiveAvgPool1d float32 [64 768 1] [64 768 1] noncontiguous bwd 13472 60656 FALSE 8249 1.633167657
AdaptiveAvgPool1d float32 [64 2048 8] [64 2048 1] contiguous fwd 34192 34192 FALSE 20569 1.662307356
AdaptiveAvgPool1d float32 [64 2048 8] [64 2048 1] contiguous bwd 13280 13280 FALSE 27751 0.4785413138
AdaptiveAvgPool1d float32 [64 2048 8] [64 2048 1] noncontiguous fwd 45264 45264 FALSE 34169 1.324709532
AdaptiveAvgPool1d float32 [64 2048 8] [64 2048 1] noncontiguous bwd 50192 98064 FALSE 27698 1.812116398
AdaptiveAvgPool1d float32 [64 72 22] [64 72 1] contiguous fwd 12976 12976 FALSE 12764 1.016609213
AdaptiveAvgPool1d float32 [64 72 22] [64 72 1] contiguous bwd 7808 7808 FALSE 6080 1.284210526
AdaptiveAvgPool1d float32 [64 72 22] [64 72 1] noncontiguous fwd 34864 34864 FALSE 19449 1.792585737
AdaptiveAvgPool1d float32 [64 72 22] [64 72 1] noncontiguous bwd 18128 60352 FALSE 6222 2.913532626
AdaptiveAvgPool1d float32 [64 120 22] [64 120 1] contiguous fwd 14496 14496 FALSE 12231 1.185185185
AdaptiveAvgPool1d float32 [64 120 22] [64 120 1] contiguous bwd 8096 8096 FALSE 7200 1.124444444
AdaptiveAvgPool1d float32 [64 120 22] [64 120 1] noncontiguous fwd 35088 35088 FALSE 19947 1.759061513
AdaptiveAvgPool1d float32 [64 120 22] [64 120 1] noncontiguous bwd 19152 62896 FALSE 7040 2.720454545
AdaptiveAvgPool1d float32 [64 480 11] [64 480 1] contiguous fwd 20832 20832 FALSE 11200 1.86
AdaptiveAvgPool1d float32 [64 480 11] [64 480 1] contiguous bwd 8944 8944 FALSE 11236 0.7960128159
AdaptiveAvgPool1d float32 [64 480 11] [64 480 1] noncontiguous fwd 25360 25360 FALSE 16836 1.506296032
AdaptiveAvgPool1d float32 [64 480 11] [64 480 1] noncontiguous bwd 23856 64432 FALSE 11538 2.067602704
AdaptiveAvgPool1d float32 [64 672 11] [64 672 1] contiguous fwd 24272 24272 FALSE 13191 1.840042453
AdaptiveAvgPool1d float32 [64 672 11] [64 672 1] contiguous bwd 9136 9136 FALSE 14151 0.6456080842
AdaptiveAvgPool1d float32 [64 672 11] [64 672 1] noncontiguous fwd 26432 26432 FALSE 19449 1.359041596
AdaptiveAvgPool1d float32 [64 672 11] [64 672 1] noncontiguous bwd 27728 78928 FALSE 14009 1.979299022
AdaptiveAvgPool1d float32 [64 960 11] [64 960 1] contiguous fwd 30320 30320 FALSE 18240 1.662280702
AdaptiveAvgPool1d float32 [64 960 11] [64 960 1] contiguous bwd 9968 9968 FALSE 18987 0.5249907832
AdaptiveAvgPool1d float32 [64 960 11] [64 960 1] noncontiguous fwd 38128 38128 FALSE 24071 1.583980724
AdaptiveAvgPool1d float32 [64 960 11] [64 960 1] noncontiguous bwd 35536 89328 FALSE 19075 1.862961992
AdaptiveAvgPool1d float32 [55 72 14] [55 72 1] contiguous fwd 11808 11808 FALSE 9084 1.2998679
AdaptiveAvgPool1d float32 [55 72 14] [55 72 1] contiguous bwd 7840 7840 FALSE 6009 1.304709602
AdaptiveAvgPool1d float32 [55 72 14] [55 72 1] noncontiguous fwd 17840 17840 FALSE 15147 1.177790982
AdaptiveAvgPool1d float32 [55 72 14] [55 72 1] noncontiguous bwd 16688 76656 FALSE 6116 2.728580772
AdaptiveAvgPool1d float32 [55 120 14] [55 120 1] contiguous fwd 12048 12048 FALSE 9031 1.334071531
AdaptiveAvgPool1d float32 [55 120 14] [55 120 1] contiguous bwd 8128 8128 FALSE 6098 1.332896031
AdaptiveAvgPool1d float32 [55 120 14] [55 120 1] noncontiguous fwd 19328 19328 FALSE 15307 1.262690272
AdaptiveAvgPool1d float32 [55 120 14] [55 120 1] noncontiguous bwd 17376 75232 FALSE 6276 2.768642447
AdaptiveAvgPool1d float32 [55 480 7] [55 480 1] contiguous fwd 13120 13120 FALSE 7538 1.740514725
AdaptiveAvgPool1d float32 [55 480 7] [55 480 1] contiguous bwd 8864 8864 FALSE 7378 1.201409596
AdaptiveAvgPool1d float32 [55 480 7] [55 480 1] noncontiguous fwd 17024 17024 FALSE 10987 1.549467553
AdaptiveAvgPool1d float32 [55 480 7] [55 480 1] noncontiguous bwd 18896 79632 FALSE 7129 2.650582129
AdaptiveAvgPool1d float32 [55 672 7] [55 672 1] contiguous fwd 14800 14800 FALSE 7893 1.875079184
AdaptiveAvgPool1d float32 [55 672 7] [55 672 1] contiguous bwd 8880 8880 FALSE 10489 0.8466012013
AdaptiveAvgPool1d float32 [55 672 7] [55 672 1] noncontiguous fwd 18368 18368 FALSE 13138 1.3980819
AdaptiveAvgPool1d float32 [55 672 7] [55 672 1] noncontiguous bwd 20848 71520 FALSE 10044 2.075667065
AdaptiveAvgPool1d float32 [55 960 7] [55 960 1] contiguous fwd 18528 18528 FALSE 8320 2.226923077
AdaptiveAvgPool1d float32 [55 960 7] [55 960 1] contiguous bwd 9264 9264 FALSE 12071 0.7674591997
AdaptiveAvgPool1d float32 [55 960 7] [55 960 1] noncontiguous fwd 20928 20928 FALSE 14524 1.440925365
AdaptiveAvgPool1d float32 [55 960 7] [55 960 1] noncontiguous bwd 24464 65008 FALSE 11840 2.066216216
AdaptiveAvgPool1d float32 [64 72 19] [64 72 1] contiguous fwd 12784 12784 FALSE 12782 1.00015647
AdaptiveAvgPool1d float32 [64 72 19] [64 72 1] contiguous bwd 7968 7968 FALSE 6009 1.326010984
AdaptiveAvgPool1d float32 [64 72 19] [64 72 1] noncontiguous fwd 32608 32608 FALSE 18062 1.805337172
AdaptiveAvgPool1d float32 [64 72 19] [64 72 1] noncontiguous bwd 18288 62368 FALSE 6151 2.973175093
AdaptiveAvgPool1d float32 [64 120 19] [64 120 1] contiguous fwd 14224 14224 FALSE 12533 1.134923801
AdaptiveAvgPool1d float32 [64 120 19] [64 120 1] contiguous bwd 8048 8048 FALSE 6364 1.264613451
AdaptiveAvgPool1d float32 [64 120 19] [64 120 1] noncontiguous fwd 32320 32320 FALSE 18311 1.765059254
AdaptiveAvgPool1d float32 [64 120 19] [64 120 1] noncontiguous bwd 18720 60928 FALSE 6596 2.838083687
AdaptiveAvgPool1d float32 [64 480 10] [64 480 1] contiguous fwd 20560 20560 FALSE 10062 2.043331346
AdaptiveAvgPool1d float32 [64 480 10] [64 480 1] contiguous bwd 8848 8848 FALSE 11147 0.7937561676
AdaptiveAvgPool1d float32 [64 480 10] [64 480 1] noncontiguous fwd 23744 23744 FALSE 15822 1.500695234
AdaptiveAvgPool1d float32 [64 480 10] [64 480 1] noncontiguous bwd 23952 70480 FALSE 11360 2.108450704
AdaptiveAvgPool1d float32 [64 672 10] [64 672 1] contiguous fwd 23792 23792 FALSE 11627 2.046271609
AdaptiveAvgPool1d float32 [64 672 10] [64 672 1] contiguous bwd 9216 9216 FALSE 13440 0.6857142857
AdaptiveAvgPool1d float32 [64 672 10] [64 672 1] noncontiguous fwd 25392 25392 FALSE 18133 1.400319859
AdaptiveAvgPool1d float32 [64 672 10] [64 672 1] noncontiguous bwd 28240 68000 FALSE 13333 2.118052951
AdaptiveAvgPool1d float32 [64 960 10] [64 960 1] contiguous fwd 29984 29984 FALSE 15413 1.945370791
AdaptiveAvgPool1d float32 [64 960 10] [64 960 1] contiguous bwd 9760 9760 FALSE 17404 0.5607906228
AdaptiveAvgPool1d float32 [64 960 10] [64 960 1] noncontiguous fwd 35408 35408 FALSE 22578 1.568252281
AdaptiveAvgPool1d float32 [64 960 10] [64 960 1] noncontiguous bwd 33872 85104 FALSE 17529 1.932340693
AdaptiveAvgPool1d float32 [64 72 25] [64 72 1] contiguous fwd 13024 13024 FALSE 16498 0.7894290217
AdaptiveAvgPool1d float32 [64 72 25] [64 72 1] contiguous bwd 7856 7856 FALSE 6204 1.266279819
AdaptiveAvgPool1d float32 [64 72 25] [64 72 1] noncontiguous fwd 38048 38048 FALSE 21920 1.735766423
AdaptiveAvgPool1d float32 [64 72 25] [64 72 1] noncontiguous bwd 18160 62304 FALSE 6276 2.893562779
AdaptiveAvgPool1d float32 [64 120 25] [64 120 1] contiguous fwd 14336 14336 FALSE 16372 0.8756413389
AdaptiveAvgPool1d float32 [64 120 25] [64 120 1] contiguous bwd 8000 8000 FALSE 7769 1.029733556
AdaptiveAvgPool1d float32 [64 120 25] [64 120 1] noncontiguous fwd 37936 37936 FALSE 21564 1.759228344
AdaptiveAvgPool1d float32 [64 120 25] [64 120 1] noncontiguous bwd 19696 59408 FALSE 7644 2.576661434
AdaptiveAvgPool1d float32 [64 480 13] [64 480 1] contiguous fwd 21872 21872 FALSE 13333 1.640441011
AdaptiveAvgPool1d float32 [64 480 13] [64 480 1] contiguous bwd 9152 9152 FALSE 12569 0.7281406635
AdaptiveAvgPool1d float32 [64 480 13] [64 480 1] noncontiguous fwd 26816 26816 FALSE 18969 1.413674943
AdaptiveAvgPool1d float32 [64 480 13] [64 480 1] noncontiguous bwd 25664 67152 FALSE 12284 2.089221752
AdaptiveAvgPool1d float32 [64 672 13] [64 672 1] contiguous fwd 24624 24624 FALSE 16391 1.502287841
AdaptiveAvgPool1d float32 [64 672 13] [64 672 1] contiguous bwd 9328 9328 FALSE 16124 0.5785164971
AdaptiveAvgPool1d float32 [64 672 13] [64 672 1] noncontiguous fwd 27952 27952 FALSE 21707 1.287695214
AdaptiveAvgPool1d float32 [64 672 13] [64 672 1] noncontiguous bwd 31152 71280 FALSE 15928 1.955801105
AdaptiveAvgPool1d float32 [64 960 13] [64 960 1] contiguous fwd 31088 31088 FALSE 24427 1.272690056
AdaptiveAvgPool1d float32 [64 960 13] [64 960 1] contiguous bwd 10224 10224 FALSE 21635 0.4725675988
AdaptiveAvgPool1d float32 [64 960 13] [64 960 1] noncontiguous fwd 41552 41552 FALSE 27342 1.519713262
AdaptiveAvgPool1d float32 [64 960 13] [64 960 1] noncontiguous bwd 39648 91120 FALSE 21884 1.811734601
AdaptiveAvgPool1d float32 [54 72 25] [54 72 1] contiguous fwd 12800 12800 FALSE 16604 0.7708985787
AdaptiveAvgPool1d float32 [54 72 25] [54 72 1] contiguous bwd 7584 7584 FALSE 6009 1.26210684
AdaptiveAvgPool1d float32 [54 72 25] [54 72 1] noncontiguous fwd 21664 21664 FALSE 19538 1.108813594
AdaptiveAvgPool1d float32 [54 72 25] [54 72 1] noncontiguous bwd 17264 65680 FALSE 6347 2.720025209
AdaptiveAvgPool1d float32 [47 72 25] [47 72 1] contiguous fwd 12400 12400 FALSE 16835 0.7365607366
AdaptiveAvgPool1d float32 [47 72 25] [47 72 1] contiguous bwd 7712 7712 FALSE 5955 1.29504618
AdaptiveAvgPool1d float32 [47 72 25] [47 72 1] noncontiguous fwd 21600 21600 FALSE 19680 1.097560976
AdaptiveAvgPool1d float32 [47 72 25] [47 72 1] noncontiguous bwd 17120 58736 FALSE 6116 2.799215173
AdaptiveAvgPool1d float32 [54 120 25] [54 120 1] contiguous fwd 13504 13504 FALSE 16427 0.8220612406
AdaptiveAvgPool1d float32 [54 120 25] [54 120 1] contiguous bwd 7872 7872 FALSE 7147 1.101441164
AdaptiveAvgPool1d float32 [54 120 25] [54 120 1] noncontiguous fwd 23536 23536 FALSE 19271 1.221317005
AdaptiveAvgPool1d float32 [54 120 25] [54 120 1] noncontiguous bwd 18928 61040 FALSE 6898 2.743983763
AdaptiveAvgPool1d float32 [47 120 25] [47 120 1] contiguous fwd 13024 13024 FALSE 16284 0.7998034881
AdaptiveAvgPool1d float32 [47 120 25] [47 120 1] contiguous bwd 8000 8000 FALSE 6382 1.253525541
AdaptiveAvgPool1d float32 [47 120 25] [47 120 1] noncontiguous fwd 24864 24864 FALSE 19555 1.271490667
AdaptiveAvgPool1d float32 [47 120 25] [47 120 1] noncontiguous bwd 18400 65856 FALSE 6364 2.891263356
AdaptiveAvgPool1d float32 [54 480 13] [54 480 1] contiguous fwd 18448 18448 FALSE 8853 2.083813397
AdaptiveAvgPool1d float32 [54 480 13] [54 480 1] contiguous bwd 8640 8640 FALSE 11200 0.7714285714
AdaptiveAvgPool1d float32 [54 480 13] [54 480 1] noncontiguous fwd 20864 20864 FALSE 15378 1.3567434
AdaptiveAvgPool1d float32 [54 480 13] [54 480 1] noncontiguous bwd 24112 66288 FALSE 11467 2.102729572
AdaptiveAvgPool1d float32 [47 480 13] [47 480 1] contiguous fwd 16816 16816 FALSE 8853 1.899469107
AdaptiveAvgPool1d float32 [47 480 13] [47 480 1] contiguous bwd 8576 8576 FALSE 10613 0.8080655799
AdaptiveAvgPool1d float32 [47 480 13] [47 480 1] noncontiguous fwd 20624 20624 FALSE 15022 1.372919718
AdaptiveAvgPool1d float32 [47 480 13] [47 480 1] noncontiguous bwd 22416 68640 FALSE 11058 2.02712968
AdaptiveAvgPool1d float32 [54 672 13] [54 672 1] contiguous fwd 22480 22480 FALSE 13689 1.642194463
AdaptiveAvgPool1d float32 [54 672 13] [54 672 1] contiguous bwd 9120 9120 FALSE 14098 0.6469002695
AdaptiveAvgPool1d float32 [54 672 13] [54 672 1] noncontiguous fwd 23600 23600 FALSE 18258 1.292584073
AdaptiveAvgPool1d float32 [54 672 13] [54 672 1] noncontiguous bwd 27984 73456 FALSE 14098 1.984962406
AdaptiveAvgPool1d float32 [47 672 13] [47 672 1] contiguous fwd 21248 21248 FALSE 13493 1.574742459
AdaptiveAvgPool1d float32 [47 672 13] [47 672 1] contiguous bwd 8848 8848 FALSE 12836 0.6893113119
AdaptiveAvgPool1d float32 [47 672 13] [47 672 1] noncontiguous fwd 22416 22416 FALSE 17315 1.294600058
AdaptiveAvgPool1d float32 [47 672 13] [47 672 1] noncontiguous bwd 25968 74191 FALSE 12747 2.03718522
AdaptiveAvgPool1d float32 [47 960 13] [47 960 1] contiguous fwd 25152 25152 FALSE 17298 1.45404093
AdaptiveAvgPool1d float32 [47 960 13] [47 960 1] contiguous bwd 9392 9392 FALSE 16800 0.559047619
AdaptiveAvgPool1d float32 [47 960 13] [47 960 1] noncontiguous fwd 26080 26080 FALSE 20409 1.277867607
AdaptiveAvgPool1d float32 [47 960 13] [47 960 1] noncontiguous bwd 32688 87536 FALSE 16853 1.939595324
AdaptiveAvgPool1d float32 [54 960 13] [54 960 1] contiguous fwd 27568 27568 FALSE 20320 1.356692913
AdaptiveAvgPool1d float32 [54 960 13] [54 960 1] contiguous bwd 9680 9680 FALSE 18987 0.5098225101
AdaptiveAvgPool1d float32 [54 960 13] [54 960 1] noncontiguous fwd 28304 28304 FALSE 22169 1.276737787
AdaptiveAvgPool1d float32 [54 960 13] [54 960 1] noncontiguous bwd 35056 89056 FALSE 19022 1.842918726
AdaptiveAvgPool1d float32 [512 1280 7] [512 1280 1] contiguous fwd 114559 114559 FALSE 52924 2.164594513
AdaptiveAvgPool1d float32 [512 1280 7] [512 1280 1] contiguous bwd 34656 34656 FALSE 111751 0.3101180303
AdaptiveAvgPool1d float32 [512 1280 7] [512 1280 1] noncontiguous fwd 175952 175952 FALSE 131697 1.336036508
AdaptiveAvgPool1d float32 [512 1280 7] [512 1280 1] noncontiguous bwd 327518 346430 FALSE 112391 2.914094545
AdaptiveAvgPool1d float32 [256 1280 7] [256 1280 1] contiguous fwd 63280 63280 FALSE 32213 1.964424301
AdaptiveAvgPool1d float32 [256 1280 7] [256 1280 1] contiguous bwd 20640 20640 FALSE 57475 0.3591126577
AdaptiveAvgPool1d float32 [256 1280 7] [256 1280 1] noncontiguous fwd 92720 92720 FALSE 68017 1.363188615
AdaptiveAvgPool1d float32 [256 1280 7] [256 1280 1] noncontiguous bwd 156367 199583 FALSE 57689 2.710516736
AdaptiveAvgPool1d float32 [256 16 56] [256 16 1] contiguous fwd 15024 15024 FALSE 19698 0.7627170271
AdaptiveAvgPool1d float32 [256 16 56] [256 16 1] contiguous bwd 8016 8016 FALSE 8711 0.9202158191
AdaptiveAvgPool1d float32 [256 16 56] [256 16 1] noncontiguous fwd 37024 37024 FALSE 32764 1.130020754
AdaptiveAvgPool1d float32 [256 16 56] [256 16 1] noncontiguous bwd 22544 63360 FALSE 8942 2.521136211
AdaptiveAvgPool1d float32 [256 96 14] [256 96 1] contiguous fwd 17888 17888 FALSE 9084 1.969176574
AdaptiveAvgPool1d float32 [256 96 14] [256 96 1] contiguous bwd 8720 8720 FALSE 11236 0.7760768957
AdaptiveAvgPool1d float32 [256 96 14] [256 96 1] noncontiguous fwd 27408 27408 FALSE 15947 1.718693171
AdaptiveAvgPool1d float32 [256 96 14] [256 96 1] noncontiguous bwd 25792 66255 FALSE 11289 2.284701922
AdaptiveAvgPool1d float32 [256 240 14] [256 240 1] contiguous fwd 31264 31264 FALSE 27182 1.150172909
AdaptiveAvgPool1d float32 [256 240 14] [256 240 1] contiguous bwd 11328 11328 FALSE 23324 0.4856799863
AdaptiveAvgPool1d float32 [256 240 14] [256 240 1] noncontiguous fwd 47200 47200 FALSE 33920 1.391509434
AdaptiveAvgPool1d float32 [256 240 14] [256 240 1] noncontiguous bwd 41232 90640 FALSE 23342 1.766429612
AdaptiveAvgPool1d float32 [256 120 14] [256 120 1] contiguous fwd 21392 21392 FALSE 14702 1.455040131
AdaptiveAvgPool1d float32 [256 120 14] [256 120 1] contiguous bwd 9008 9008 FALSE 13298 0.677395097
AdaptiveAvgPool1d float32 [256 120 14] [256 120 1] noncontiguous fwd 28768 28768 FALSE 22311 1.289408812
AdaptiveAvgPool1d float32 [256 120 14] [256 120 1] noncontiguous bwd 27248 84416 FALSE 13280 2.051807229
AdaptiveAvgPool1d float32 [256 144 14] [256 144 1] contiguous fwd 22880 22880 FALSE 14969 1.528492217
AdaptiveAvgPool1d float32 [256 144 14] [256 144 1] contiguous bwd 9216 9216 FALSE 15218 0.6055986332
AdaptiveAvgPool1d float32 [256 144 14] [256 144 1] noncontiguous fwd 29344 29344 FALSE 23075 1.271679307
AdaptiveAvgPool1d float32 [256 144 14] [256 144 1] noncontiguous bwd 29904 84688 FALSE 15147 1.974252327
AdaptiveAvgPool1d float32 [256 288 7] [256 288 1] contiguous fwd 22624 22624 FALSE 11075 2.042799097
AdaptiveAvgPool1d float32 [256 288 7] [256 288 1] contiguous bwd 9552 9552 FALSE 15307 0.6240282224
AdaptiveAvgPool1d float32 [256 288 7] [256 288 1] noncontiguous fwd 31184 31184 FALSE 20693 1.506983038
AdaptiveAvgPool1d float32 [256 288 7] [256 288 1] noncontiguous bwd 30528 81680 FALSE 15733 1.940380093
AdaptiveAvgPool1d float32 [256 576 7] [256 576 1] contiguous fwd 35216 35216 FALSE 17511 2.01107875
AdaptiveAvgPool1d float32 [256 576 7] [256 576 1] contiguous bwd 13456 13456 FALSE 27307 0.4927674223
AdaptiveAvgPool1d float32 [256 576 7] [256 576 1] noncontiguous fwd 46160 46160 FALSE 35502 1.300208439
AdaptiveAvgPool1d float32 [256 576 7] [256 576 1] noncontiguous bwd 50288 101407 FALSE 27449 1.832052169
AdaptiveAvgPool1d float32 [256 2048 7] [256 2048 1] contiguous fwd 93535 93535 FALSE 44836 2.086158444
AdaptiveAvgPool1d float32 [256 2048 7] [256 2048 1] contiguous bwd 29008 29008 FALSE 89831 0.3229174784
AdaptiveAvgPool1d float32 [256 2048 7] [256 2048 1] noncontiguous fwd 151695 151695 FALSE 104941 1.445526534
AdaptiveAvgPool1d float32 [256 2048 7] [256 2048 1] noncontiguous bwd 349838 374798 FALSE 90364 3.871431101
AdaptiveAvgPool1d float32 [1536 100 13] [1536 100 1] contiguous fwd 58512 58512 FALSE 49582 1.180105684
AdaptiveAvgPool1d float32 [1536 100 13] [1536 100 1] contiguous bwd 17760 17760 FALSE 50151 0.3541305258
AdaptiveAvgPool1d float32 [1536 100 13] [1536 100 1] noncontiguous fwd 69472 69472 FALSE 55182 1.258961255
AdaptiveAvgPool1d float32 [1536 100 13] [1536 100 1] noncontiguous bwd 84288 126559 FALSE 50435 1.671220383
AdaptiveAvgPool1d float32 [731 100 13] [731 100 1] contiguous fwd 34080 34080 FALSE 26631 1.279711614
AdaptiveAvgPool1d float32 [731 100 13] [731 100 1] contiguous bwd 12000 12000 FALSE 25333 0.4736904433
AdaptiveAvgPool1d float32 [731 100 13] [731 100 1] noncontiguous fwd 30032 30032 FALSE 16391 1.832225002
AdaptiveAvgPool1d float32 [731 100 13] [731 100 1] noncontiguous bwd 44816 92143 FALSE 25280 1.77278481
AdaptiveAvgPool1d float32 [724 100 13] [724 100 1] contiguous fwd 33920 33920 FALSE 26809 1.265246745
AdaptiveAvgPool1d float32 [724 100 13] [724 100 1] contiguous bwd 11856 11856 FALSE 25298 0.4686536485
AdaptiveAvgPool1d float32 [724 100 13] [724 100 1] noncontiguous fwd 31040 31040 FALSE 16604 1.869429053
AdaptiveAvgPool1d float32 [724 100 13] [724 100 1] noncontiguous bwd 44240 92512 FALSE 25262 1.751246932
AdaptiveAvgPool1d float32 [64 16 80] [64 16 1] contiguous fwd 12672 12672 FALSE 24338 0.5206672693
AdaptiveAvgPool1d float32 [64 16 80] [64 16 1] contiguous bwd 6752 6752 FALSE 4836 1.396195203
AdaptiveAvgPool1d float32 [64 16 80] [64 16 1] noncontiguous fwd 50544 50544 FALSE 44711 1.130460066
AdaptiveAvgPool1d float32 [64 16 80] [64 16 1] noncontiguous bwd 17280 60096 FALSE 4978 3.471273604
AdaptiveAvgPool1d float32 [64 96 20] [64 96 1] contiguous fwd 13072 13072 FALSE 11004 1.187931661
AdaptiveAvgPool1d float32 [64 96 20] [64 96 1] contiguous bwd 8000 8000 FALSE 6187 1.293033781
AdaptiveAvgPool1d float32 [64 96 20] [64 96 1] noncontiguous fwd 31680 31680 FALSE 18524 1.710213777
AdaptiveAvgPool1d float32 [64 96 20] [64 96 1] noncontiguous bwd 18480 62288 FALSE 6222 2.970106075
AdaptiveAvgPool1d float32 [64 240 20] [64 240 1] contiguous fwd 20688 20688 FALSE 10933 1.892252813
AdaptiveAvgPool1d float32 [64 240 20] [64 240 1] contiguous bwd 8448 8448 FALSE 10773 0.7841826789
AdaptiveAvgPool1d float32 [64 240 20] [64 240 1] noncontiguous fwd 33568 33568 FALSE 19680 1.705691057
AdaptiveAvgPool1d float32 [64 240 20] [64 240 1] noncontiguous bwd 24000 66160 FALSE 10809 2.220371912
AdaptiveAvgPool1d float32 [64 120 20] [64 120 1] contiguous fwd 14208 14208 FALSE 11200 1.268571429
AdaptiveAvgPool1d float32 [64 120 20] [64 120 1] contiguous bwd 8032 8032 FALSE 6524 1.231146536
AdaptiveAvgPool1d float32 [64 120 20] [64 120 1] noncontiguous fwd 32560 32560 FALSE 18773 1.734405796
AdaptiveAvgPool1d float32 [64 120 20] [64 120 1] noncontiguous bwd 18752 71424 FALSE 6524 2.874310239
AdaptiveAvgPool1d float32 [64 144 20] [64 144 1] contiguous fwd 14880 14880 FALSE 10738 1.385732911
AdaptiveAvgPool1d float32 [64 144 20] [64 144 1] contiguous bwd 8144 8144 FALSE 7289 1.117300041
AdaptiveAvgPool1d float32 [64 144 20] [64 144 1] noncontiguous fwd 32240 32240 FALSE 19040 1.693277311
AdaptiveAvgPool1d float32 [64 144 20] [64 144 1] noncontiguous bwd 20624 71744 FALSE 7076 2.91464104
AdaptiveAvgPool1d float32 [64 288 10] [64 288 1] contiguous fwd 14768 14768 FALSE 8249 1.790277609
AdaptiveAvgPool1d float32 [64 288 10] [64 288 1] contiguous bwd 8608 8608 FALSE 7360 1.169565217
AdaptiveAvgPool1d float32 [64 288 10] [64 288 1] noncontiguous fwd 23392 23392 FALSE 13760 1.7
AdaptiveAvgPool1d float32 [64 288 10] [64 288 1] noncontiguous bwd 20464 60720 FALSE 7129 2.870528826
AdaptiveAvgPool1d float32 [256 512 7] [256 512 7] contiguous fwd 226367 226367 FALSE 20071 11.27831199
AdaptiveAvgPool1d float32 [256 512 7] [256 512 7] contiguous bwd 216367 232962 FALSE 24906 8.687344415
AdaptiveAvgPool1d float32 [256 512 7] [256 512 7] noncontiguous fwd 257231 274143 FALSE 33422 7.696457423
AdaptiveAvgPool1d float32 [256 512 7] [256 512 7] noncontiguous bwd 249855 283647 FALSE 33369 7.487638227
AdaptiveAvgPool1d bfloat16 [256 256 6] [256 256 6] contiguous fwd 117935 117935 FALSE 11182 10.54686103
AdaptiveAvgPool1d bfloat16 [256 256 6] [256 256 6] contiguous bwd 138271 153951 FALSE 12782 10.81763417
AdaptiveAvgPool1d bfloat16 [256 256 6] [256 256 6] noncontiguous fwd 135119 151119 FALSE 18098 7.46596309
AdaptiveAvgPool1d bfloat16 [256 256 6] [256 256 6] noncontiguous bwd 155488 186815 FALSE 18364 8.467000653
AdaptiveAvgPool1d bfloat16 [16 72 64] [16 72 1] contiguous fwd 12064 12064 FALSE 28284 0.4265309009
AdaptiveAvgPool1d bfloat16 [16 72 64] [16 72 1] contiguous bwd 6320 6320 FALSE 4498 1.40506892
AdaptiveAvgPool1d bfloat16 [16 72 64] [16 72 1] noncontiguous fwd 29728 29728 FALSE 34471 0.8624060805
AdaptiveAvgPool1d bfloat16 [16 72 64] [16 72 1] noncontiguous bwd 15056 55632 FALSE 4658 3.232288536
AdaptiveAvgPool1d bfloat16 [16 120 64] [16 120 1] contiguous fwd 12816 12816 FALSE 27875 0.4597668161
AdaptiveAvgPool1d bfloat16 [16 120 64] [16 120 1] contiguous bwd 6816 6816 FALSE 5689 1.1981016
AdaptiveAvgPool1d bfloat16 [16 120 64] [16 120 1] noncontiguous fwd 29728 29728 FALSE 33120 0.8975845411
AdaptiveAvgPool1d bfloat16 [16 120 64] [16 120 1] noncontiguous bwd 16576 54144 FALSE 5636 2.941092974
AdaptiveAvgPool1d bfloat16 [16 480 32] [16 480 1] contiguous fwd 17008 17008 FALSE 12338 1.37850543
AdaptiveAvgPool1d bfloat16 [16 480 32] [16 480 1] contiguous bwd 8192 8192 FALSE 9138 0.896476253
AdaptiveAvgPool1d bfloat16 [16 480 32] [16 480 1] noncontiguous fwd 21488 21488 FALSE 21102 1.018292105
AdaptiveAvgPool1d bfloat16 [16 480 32] [16 480 1] noncontiguous bwd 20768 57936 FALSE 10133 2.049541103
AdaptiveAvgPool1d bfloat16 [16 672 32] [16 672 1] contiguous fwd 18720 18720 FALSE 12160 1.539473684
AdaptiveAvgPool1d bfloat16 [16 672 32] [16 672 1] contiguous bwd 8624 8624 FALSE 11164 0.772482981
AdaptiveAvgPool1d bfloat16 [16 672 32] [16 672 1] noncontiguous fwd 21760 21760 FALSE 21440 1.014925373
AdaptiveAvgPool1d bfloat16 [16 672 32] [16 672 1] noncontiguous bwd 24864 67568 FALSE 11271 2.206015438
AdaptiveAvgPool1d bfloat16 [16 960 32] [16 960 1] contiguous fwd 22784 22784 FALSE 12640 1.802531646
AdaptiveAvgPool1d bfloat16 [16 960 32] [16 960 1] contiguous bwd 9280 9280 FALSE 14951 0.6206942679
AdaptiveAvgPool1d bfloat16 [16 960 32] [16 960 1] noncontiguous fwd 22096 22096 FALSE 21973 1.005597779
AdaptiveAvgPool1d bfloat16 [16 960 32] [16 960 1] noncontiguous bwd 28992 66704 FALSE 14791 1.960110878
AdaptiveAvgPool1d bfloat16 [3 2048 64] [3 2048 1] contiguous fwd 20016 20016 FALSE 27627 0.7245086329
AdaptiveAvgPool1d bfloat16 [3 2048 64] [3 2048 1] contiguous bwd 8752 8752 FALSE 12320 0.7103896104
AdaptiveAvgPool1d bfloat16 [3 2048 64] [3 2048 1] noncontiguous fwd 21008 21008 FALSE 32924 0.638075568
AdaptiveAvgPool1d bfloat16 [3 2048 64] [3 2048 1] noncontiguous bwd 25776 78800 FALSE 12284 2.098339303
AdaptiveAvgPool1d bfloat16 [64 2208 7] [64 2208 1] contiguous fwd 28928 28928 FALSE 10507 2.753212144
AdaptiveAvgPool1d bfloat16 [64 2208 7] [64 2208 1] contiguous bwd 12272 12272 FALSE 27076 0.4532427242
AdaptiveAvgPool1d bfloat16 [64 2208 7] [64 2208 1] noncontiguous fwd 40688 40688 FALSE 29333 1.387106672
AdaptiveAvgPool1d bfloat16 [64 2208 7] [64 2208 1] noncontiguous bwd 39824 88320 FALSE 27235 1.462236093
AdaptiveAvgPool1d bfloat16 [256 512 14] [256 512 4] contiguous fwd 255839 255839 FALSE 17333 14.76022616
AdaptiveAvgPool1d bfloat16 [256 512 14] [256 512 4] contiguous bwd 363822 381102 FALSE 61813 5.885849255
AdaptiveAvgPool1d bfloat16 [256 512 14] [256 512 4] noncontiguous fwd 309407 326239 FALSE 57831 5.350192803
AdaptiveAvgPool1d bfloat16 [256 512 14] [256 512 4] noncontiguous bwd 419134 453902 FALSE 62506 6.705500272
AdaptiveAvgPool1d bfloat16 [256 528 14] [256 528 4] contiguous fwd 263103 263103 FALSE 17671 14.88897063
AdaptiveAvgPool1d bfloat16 [256 528 14] [256 528 4] contiguous bwd 374622 391710 FALSE 63893 5.863271407
AdaptiveAvgPool1d bfloat16 [256 528 14] [256 528 4] noncontiguous fwd 318479 335103 FALSE 60658 5.250403904
AdaptiveAvgPool1d bfloat16 [256 528 14] [256 528 4] noncontiguous bwd 429854 464398 FALSE 64657 6.64822061
AdaptiveAvgPool1d bfloat16 [256 1024 7] [256 1024 1] contiguous fwd 43984 43984 FALSE 17618 2.496537632
AdaptiveAvgPool1d bfloat16 [256 1024 7] [256 1024 1] contiguous bwd 16288 16288 FALSE 47680 0.3416107383
AdaptiveAvgPool1d bfloat16 [256 1024 7] [256 1024 1] noncontiguous fwd 69296 69296 FALSE 55484 1.24893663
AdaptiveAvgPool1d bfloat16 [256 1024 7] [256 1024 1] noncontiguous bwd 71520 110095 FALSE 47573 1.503373762
AdaptiveAvgPool1d bfloat16 [64 768 1] [64 768 1] contiguous fwd 12256 12256 FALSE 6169 1.986707732
AdaptiveAvgPool1d bfloat16 [64 768 1] [64 768 1] contiguous bwd 5584 5584 FALSE 6204 0.9000644745
AdaptiveAvgPool1d bfloat16 [64 768 1] [64 768 1] noncontiguous fwd 12816 12816 FALSE 7004 1.829811536
AdaptiveAvgPool1d bfloat16 [64 768 1] [64 768 1] noncontiguous bwd 14032 59792 FALSE 7040 1.993181818
AdaptiveAvgPool1d bfloat16 [64 2048 8] [64 2048 1] contiguous fwd 34816 34816 FALSE 13902 2.504387858
AdaptiveAvgPool1d bfloat16 [64 2048 8] [64 2048 1] contiguous bwd 12608 12608 FALSE 28356 0.4446325293
AdaptiveAvgPool1d bfloat16 [64 2048 8] [64 2048 1] noncontiguous fwd 38880 38880 FALSE 29635 1.311962207
AdaptiveAvgPool1d bfloat16 [64 2048 8] [64 2048 1] noncontiguous bwd 45952 87808 FALSE 28409 1.617515576
AdaptiveAvgPool1d bfloat16 [64 72 22] [64 72 1] contiguous fwd 12816 12816 FALSE 10347 1.23861989
AdaptiveAvgPool1d bfloat16 [64 72 22] [64 72 1] contiguous bwd 7744 7744 FALSE 5973 1.296500921
AdaptiveAvgPool1d bfloat16 [64 72 22] [64 72 1] noncontiguous fwd 25184 25184 FALSE 16764 1.502266762
AdaptiveAvgPool1d bfloat16 [64 72 22] [64 72 1] noncontiguous bwd 18176 63952 FALSE 6347 2.863715141
AdaptiveAvgPool1d bfloat16 [64 120 22] [64 120 1] contiguous fwd 13968 13968 FALSE 9884 1.413193039
AdaptiveAvgPool1d bfloat16 [64 120 22] [64 120 1] contiguous bwd 8128 8128 FALSE 7271 1.117865493
AdaptiveAvgPool1d bfloat16 [64 120 22] [64 120 1] noncontiguous fwd 26304 26304 FALSE 17351 1.515993315
AdaptiveAvgPool1d bfloat16 [64 120 22] [64 120 1] noncontiguous bwd 18704 62672 FALSE 7147 2.617042116
AdaptiveAvgPool1d bfloat16 [64 480 11] [64 480 1] contiguous fwd 19072 19072 FALSE 8373 2.277797683
AdaptiveAvgPool1d bfloat16 [64 480 11] [64 480 1] contiguous bwd 8656 8656 FALSE 11271 0.7679886434
AdaptiveAvgPool1d bfloat16 [64 480 11] [64 480 1] noncontiguous fwd 21248 21248 FALSE 16070 1.322215308
AdaptiveAvgPool1d bfloat16 [64 480 11] [64 480 1] noncontiguous bwd 21328 67487 FALSE 11680 1.826027397
AdaptiveAvgPool1d bfloat16 [64 672 11] [64 672 1] contiguous fwd 21376 21376 FALSE 8604 2.484425848
AdaptiveAvgPool1d bfloat16 [64 672 11] [64 672 1] contiguous bwd 9168 9168 FALSE 14436 0.6350789692
AdaptiveAvgPool1d bfloat16 [64 672 11] [64 672 1] noncontiguous fwd 24000 24000 FALSE 17867 1.343258521
AdaptiveAvgPool1d bfloat16 [64 672 11] [64 672 1] noncontiguous bwd 24240 73104 FALSE 14240 1.702247191
AdaptiveAvgPool1d bfloat16 [64 960 11] [64 960 1] contiguous fwd 27104 27104 FALSE 9884 2.742209632
AdaptiveAvgPool1d bfloat16 [64 960 11] [64 960 1] contiguous bwd 9920 9920 FALSE 19467 0.5095803154
AdaptiveAvgPool1d bfloat16 [64 960 11] [64 960 1] noncontiguous fwd 34080 34080 FALSE 23538 1.447871527
AdaptiveAvgPool1d bfloat16 [64 960 11] [64 960 1] noncontiguous bwd 31936 81392 FALSE 19431 1.643559261
AdaptiveAvgPool1d bfloat16 [55 72 14] [55 72 1] contiguous fwd 11872 11872 FALSE 9298 1.276833728
AdaptiveAvgPool1d bfloat16 [55 72 14] [55 72 1] contiguous bwd 7696 7696 FALSE 5760 1.336111111
AdaptiveAvgPool1d bfloat16 [55 72 14] [55 72 1] noncontiguous fwd 17280 17280 FALSE 14204 1.216558716
AdaptiveAvgPool1d bfloat16 [55 72 14] [55 72 1] noncontiguous bwd 16032 65504 FALSE 6027 2.660029866
AdaptiveAvgPool1d bfloat16 [55 120 14] [55 120 1] contiguous fwd 12112 12112 FALSE 8604 1.407717341
AdaptiveAvgPool1d bfloat16 [55 120 14] [55 120 1] contiguous bwd 7760 7760 FALSE 6169 1.257902415
AdaptiveAvgPool1d bfloat16 [55 120 14] [55 120 1] noncontiguous fwd 19056 19056 FALSE 14791 1.288351024
AdaptiveAvgPool1d bfloat16 [55 120 14] [55 120 1] noncontiguous bwd 16624 66048 FALSE 6436 2.582970789
AdaptiveAvgPool1d bfloat16 [55 480 7] [55 480 1] contiguous fwd 12816 12816 FALSE 7271 1.762618622
AdaptiveAvgPool1d bfloat16 [55 480 7] [55 480 1] contiguous bwd 8432 8432 FALSE 7396 1.140075717
AdaptiveAvgPool1d bfloat16 [55 480 7] [55 480 1] noncontiguous fwd 16992 16992 FALSE 10915 1.556756757
AdaptiveAvgPool1d bfloat16 [55 480 7] [55 480 1] noncontiguous bwd 17840 88848 FALSE 7253 2.45967186
AdaptiveAvgPool1d bfloat16 [55 672 7] [55 672 1] contiguous fwd 14080 14080 FALSE 7502 1.876832845
AdaptiveAvgPool1d bfloat16 [55 672 7] [55 672 1] contiguous bwd 8656 8656 FALSE 10276 0.8423511094
AdaptiveAvgPool1d bfloat16 [55 672 7] [55 672 1] noncontiguous fwd 18176 18176 FALSE 12427 1.462621711
AdaptiveAvgPool1d bfloat16 [55 672 7] [55 672 1] noncontiguous bwd 19296 66096 FALSE 10507 1.836489959
AdaptiveAvgPool1d bfloat16 [55 960 7] [55 960 1] contiguous fwd 16528 16528 FALSE 7964 2.075339026
AdaptiveAvgPool1d bfloat16 [55 960 7] [55 960 1] contiguous bwd 8992 8992 FALSE 12107 0.7427108284
AdaptiveAvgPool1d bfloat16 [55 960 7] [55 960 1] noncontiguous fwd 20880 20880 FALSE 14613 1.42886471
AdaptiveAvgPool1d bfloat16 [55 960 7] [55 960 1] noncontiguous bwd 21360 69392 FALSE 11982 1.782674011
AdaptiveAvgPool1d bfloat16 [64 72 19] [64 72 1] contiguous fwd 13088 13088 FALSE 9973 1.312343327
AdaptiveAvgPool1d bfloat16 [64 72 19] [64 72 1] contiguous bwd 7760 7760 FALSE 5956 1.302887844
AdaptiveAvgPool1d bfloat16 [64 72 19] [64 72 1] noncontiguous fwd 23792 23792 FALSE 15591 1.526008595
AdaptiveAvgPool1d bfloat16 [64 72 19] [64 72 1] noncontiguous bwd 17824 63344 FALSE 6293 2.832353409
AdaptiveAvgPool1d bfloat16 [64 120 19] [64 120 1] contiguous fwd 13856 13856 FALSE 9760 1.419672131
AdaptiveAvgPool1d bfloat16 [64 120 19] [64 120 1] contiguous bwd 7872 7872 FALSE 6471 1.216504404
AdaptiveAvgPool1d bfloat16 [64 120 19] [64 120 1] noncontiguous fwd 24688 24688 FALSE 16089 1.534464541
AdaptiveAvgPool1d bfloat16 [64 120 19] [64 120 1] noncontiguous bwd 18240 62287 FALSE 6613 2.758203538
AdaptiveAvgPool1d bfloat16 [64 480 10] [64 480 1] contiguous fwd 18736 18736 FALSE 8160 2.296078431
AdaptiveAvgPool1d bfloat16 [64 480 10] [64 480 1] contiguous bwd 8688 8688 FALSE 11076 0.7843986999
AdaptiveAvgPool1d bfloat16 [64 480 10] [64 480 1] noncontiguous fwd 20096 20096 FALSE 14933 1.345744325
AdaptiveAvgPool1d bfloat16 [64 480 10] [64 480 1] noncontiguous bwd 21424 63392 FALSE 11200 1.912857143
AdaptiveAvgPool1d bfloat16 [64 672 10] [64 672 1] contiguous fwd 21376 21376 FALSE 8231 2.597011299
AdaptiveAvgPool1d bfloat16 [64 672 10] [64 672 1] contiguous bwd 9008 9008 FALSE 13618 0.6614774563
AdaptiveAvgPool1d bfloat16 [64 672 10] [64 672 1] noncontiguous fwd 22480 22480 FALSE 16871 1.332463991
AdaptiveAvgPool1d bfloat16 [64 672 10] [64 672 1] noncontiguous bwd 24640 64896 FALSE 13564 1.816573282
AdaptiveAvgPool1d bfloat16 [64 960 10] [64 960 1] contiguous fwd 26688 26688 FALSE 9049 2.949276163
AdaptiveAvgPool1d bfloat16 [64 960 10] [64 960 1] contiguous bwd 9728 9728 FALSE 17920 0.5428571429
AdaptiveAvgPool1d bfloat16 [64 960 10] [64 960 1] noncontiguous fwd 31664 31664 FALSE 21991 1.439861762
AdaptiveAvgPool1d bfloat16 [64 960 10] [64 960 1] noncontiguous bwd 32688 68400 FALSE 17724 1.844278944
AdaptiveAvgPool1d bfloat16 [64 72 25] [64 72 1] contiguous fwd 12960 12960 FALSE 10578 1.225184345
AdaptiveAvgPool1d bfloat16 [64 72 25] [64 72 1] contiguous bwd 7712 7712 FALSE 5956 1.294828744
AdaptiveAvgPool1d bfloat16 [64 72 25] [64 72 1] noncontiguous fwd 27168 27168 FALSE 18347 1.48078705
AdaptiveAvgPool1d bfloat16 [64 72 25] [64 72 1] noncontiguous bwd 18400 61248 FALSE 6364 2.891263356
AdaptiveAvgPool1d bfloat16 [64 120 25] [64 120 1] contiguous fwd 13936 13936 FALSE 10329 1.349210959
AdaptiveAvgPool1d bfloat16 [64 120 25] [64 120 1] contiguous bwd 8080 8080 FALSE 7875 1.026031746
AdaptiveAvgPool1d bfloat16 [64 120 25] [64 120 1] noncontiguous fwd 27920 27920 FALSE 18898 1.477405016
AdaptiveAvgPool1d bfloat16 [64 120 25] [64 120 1] noncontiguous bwd 18864 60704 FALSE 7840 2.406122449
AdaptiveAvgPool1d bfloat16 [64 480 13] [64 480 1] contiguous fwd 19008 19008 FALSE 8729 2.177569023
AdaptiveAvgPool1d bfloat16 [64 480 13] [64 480 1] contiguous bwd 8928 8928 FALSE 12676 0.7043231303
AdaptiveAvgPool1d bfloat16 [64 480 13] [64 480 1] noncontiguous fwd 22256 22256 FALSE 17902 1.243213049
AdaptiveAvgPool1d bfloat16 [64 480 13] [64 480 1] noncontiguous bwd 23024 64464 FALSE 12604 1.826721676
AdaptiveAvgPool1d bfloat16 [64 672 13] [64 672 1] contiguous fwd 21712 21712 FALSE 9031 2.404163437
AdaptiveAvgPool1d bfloat16 [64 672 13] [64 672 1] contiguous bwd 9328 9328 FALSE 16462 0.5666383186
AdaptiveAvgPool1d bfloat16 [64 672 13] [64 672 1] noncontiguous fwd 25632 25632 FALSE 20249 1.265840288
AdaptiveAvgPool1d bfloat16 [64 672 13] [64 672 1] noncontiguous bwd 27568 63792 FALSE 16284 1.692950135
AdaptiveAvgPool1d bfloat16 [64 960 13] [64 960 1] contiguous fwd 27232 27232 FALSE 11111 2.450904509
AdaptiveAvgPool1d bfloat16 [64 960 13] [64 960 1] contiguous bwd 10224 10224 FALSE 22116 0.462289745
AdaptiveAvgPool1d bfloat16 [64 960 13] [64 960 1] noncontiguous fwd 37296 37296 FALSE 26916 1.385644226
AdaptiveAvgPool1d bfloat16 [64 960 13] [64 960 1] noncontiguous bwd 35824 78288 FALSE 22098 1.621142185
AdaptiveAvgPool1d bfloat16 [54 72 25] [54 72 1] contiguous fwd 12768 12768 FALSE 10435 1.223574509
AdaptiveAvgPool1d bfloat16 [54 72 25] [54 72 1] contiguous bwd 7680 7680 FALSE 5938 1.293364769
AdaptiveAvgPool1d bfloat16 [54 72 25] [54 72 1] noncontiguous fwd 21456 21456 FALSE 19609 1.094191443
AdaptiveAvgPool1d bfloat16 [54 72 25] [54 72 1] noncontiguous bwd 16592 60608 FALSE 6133 2.705364422
AdaptiveAvgPool1d bfloat16 [47 72 25] [47 72 1] contiguous fwd 12448 12448 FALSE 10880 1.144117647
AdaptiveAvgPool1d bfloat16 [47 72 25] [47 72 1] contiguous bwd 7440 7440 FALSE 5564 1.337167505
AdaptiveAvgPool1d bfloat16 [47 72 25] [47 72 1] noncontiguous fwd 21360 21360 FALSE 20000 1.068
AdaptiveAvgPool1d bfloat16 [47 72 25] [47 72 1] noncontiguous bwd 16432 54160 FALSE 5938 2.767261704
AdaptiveAvgPool1d bfloat16 [54 120 25] [54 120 1] contiguous fwd 12944 12944 FALSE 10347 1.250990625
AdaptiveAvgPool1d bfloat16 [54 120 25] [54 120 1] contiguous bwd 7872 7872 FALSE 7200 1.093333333
AdaptiveAvgPool1d bfloat16 [54 120 25] [54 120 1] noncontiguous fwd 24192 24192 FALSE 19982 1.210689621
AdaptiveAvgPool1d bfloat16 [54 120 25] [54 120 1] noncontiguous bwd 18192 56112 FALSE 7093 2.56478218
AdaptiveAvgPool1d bfloat16 [47 120 25] [47 120 1] contiguous fwd 13056 13056 FALSE 10347 1.261815019
AdaptiveAvgPool1d bfloat16 [47 120 25] [47 120 1] contiguous bwd 7824 7824 FALSE 6435 1.215850816
AdaptiveAvgPool1d bfloat16 [47 120 25] [47 120 1] noncontiguous fwd 24384 24384 FALSE 19822 1.23014832
AdaptiveAvgPool1d bfloat16 [47 120 25] [47 120 1] noncontiguous bwd 18096 53264 FALSE 6524 2.77375843
AdaptiveAvgPool1d bfloat16 [54 480 13] [54 480 1] contiguous fwd 16592 16592 FALSE 8444 1.964945523
AdaptiveAvgPool1d bfloat16 [54 480 13] [54 480 1] contiguous bwd 8480 8480 FALSE 11378 0.7452979434
AdaptiveAvgPool1d bfloat16 [54 480 13] [54 480 1] noncontiguous fwd 20640 20640 FALSE 15413 1.339129306
AdaptiveAvgPool1d bfloat16 [54 480 13] [54 480 1] noncontiguous bwd 21376 56960 FALSE 11609 1.841330003
AdaptiveAvgPool1d bfloat16 [47 480 13] [47 480 1] contiguous fwd 15616 15616 FALSE 8409 1.857057914
AdaptiveAvgPool1d bfloat16 [47 480 13] [47 480 1] contiguous bwd 8624 8624 FALSE 10507 0.8207861426
AdaptiveAvgPool1d bfloat16 [47 480 13] [47 480 1] noncontiguous fwd 20416 20416 FALSE 15360 1.329166667
AdaptiveAvgPool1d bfloat16 [47 480 13] [47 480 1] noncontiguous bwd 20240 59664 FALSE 10898 1.857221509
AdaptiveAvgPool1d bfloat16 [54 672 13] [54 672 1] contiguous fwd 20128 20128 FALSE 8853 2.273579578
AdaptiveAvgPool1d bfloat16 [54 672 13] [54 672 1] contiguous bwd 8928 8928 FALSE 14489 0.6161915936
AdaptiveAvgPool1d bfloat16 [54 672 13] [54 672 1] noncontiguous fwd 23552 23552 FALSE 18293 1.287487017
AdaptiveAvgPool1d bfloat16 [54 672 13] [54 672 1] noncontiguous bwd 23968 57744 FALSE 14311 1.674795612
AdaptiveAvgPool1d bfloat16 [47 672 13] [47 672 1] contiguous fwd 19104 19104 FALSE 8640 2.211111111
AdaptiveAvgPool1d bfloat16 [47 672 13] [47 672 1] contiguous bwd 8928 8928 FALSE 13120 0.6804878049
AdaptiveAvgPool1d bfloat16 [47 672 13] [47 672 1] noncontiguous fwd 22432 22432 FALSE 17458 1.284912361
AdaptiveAvgPool1d bfloat16 [47 672 13] [47 672 1] noncontiguous bwd 22752 57216 FALSE 12996 1.750692521
AdaptiveAvgPool1d bfloat16 [47 960 13] [47 960 1] contiguous fwd 22272 22272 FALSE 9084 2.451783355
AdaptiveAvgPool1d bfloat16 [47 960 13] [47 960 1] contiguous bwd 9408 9408 FALSE 17155 0.5484115418
AdaptiveAvgPool1d bfloat16 [47 960 13] [47 960 1] noncontiguous fwd 26176 26176 FALSE 20391 1.283703595
AdaptiveAvgPool1d bfloat16 [47 960 13] [47 960 1] noncontiguous bwd 27152 63184 FALSE 17067 1.590906428
AdaptiveAvgPool1d bfloat16 [54 960 13] [54 960 1] contiguous fwd 24144 24144 FALSE 9156 2.636959371
AdaptiveAvgPool1d bfloat16 [54 960 13] [54 960 1] contiguous bwd 9664 9664 FALSE 19431 0.497349596
AdaptiveAvgPool1d bfloat16 [54 960 13] [54 960 1] noncontiguous fwd 27984 27984 FALSE 22347 1.252248624
AdaptiveAvgPool1d bfloat16 [54 960 13] [54 960 1] noncontiguous bwd 30208 69328 FALSE 19342 1.561782649
AdaptiveAvgPool1d bfloat16 [512 1280 7] [512 1280 1] contiguous fwd 90447 90447 FALSE 26542 3.407693467
AdaptiveAvgPool1d bfloat16 [512 1280 7] [512 1280 1] contiguous bwd 32176 32176 FALSE 114417 0.281216952
AdaptiveAvgPool1d bfloat16 [512 1280 7] [512 1280 1] noncontiguous fwd 156943 156943 FALSE 126062 1.244966762
AdaptiveAvgPool1d bfloat16 [512 1280 7] [512 1280 1] noncontiguous bwd 171439 192527 FALSE 115110 1.489349318
AdaptiveAvgPool1d bfloat16 [256 1280 7] [256 1280 1] contiguous fwd 51200 51200 FALSE 18684 2.740312567
AdaptiveAvgPool1d bfloat16 [256 1280 7] [256 1280 1] contiguous bwd 18672 18672 FALSE 58702 0.3180811557
AdaptiveAvgPool1d bfloat16 [256 1280 7] [256 1280 1] noncontiguous fwd 87200 87200 FALSE 64213 1.357980471
AdaptiveAvgPool1d bfloat16 [256 1280 7] [256 1280 1] noncontiguous bwd 86032 116351 FALSE 59093 1.455874638
AdaptiveAvgPool1d bfloat16 [256 16 56] [256 16 1] contiguous fwd 14496 14496 FALSE 18453 0.785563323
AdaptiveAvgPool1d bfloat16 [256 16 56] [256 16 1] contiguous bwd 8080 8080 FALSE 8658 0.9332409332
AdaptiveAvgPool1d bfloat16 [256 16 56] [256 16 1] noncontiguous fwd 32784 32784 FALSE 32106 1.021117548
AdaptiveAvgPool1d bfloat16 [256 16 56] [256 16 1] noncontiguous bwd 22000 66080 FALSE 9013 2.440918673
AdaptiveAvgPool1d bfloat16 [256 96 14] [256 96 1] contiguous fwd 16096 16096 FALSE 8729 1.843968381
AdaptiveAvgPool1d bfloat16 [256 96 14] [256 96 1] contiguous bwd 8656 8656 FALSE 11360 0.761971831
AdaptiveAvgPool1d bfloat16 [256 96 14] [256 96 1] noncontiguous fwd 26720 26720 FALSE 15520 1.721649485
AdaptiveAvgPool1d bfloat16 [256 96 14] [256 96 1] noncontiguous bwd 24144 63936 FALSE 11680 2.067123288
AdaptiveAvgPool1d bfloat16 [256 240 14] [256 240 1] contiguous fwd 27344 27344 FALSE 12195 2.242230422
AdaptiveAvgPool1d bfloat16 [256 240 14] [256 240 1] contiguous bwd 11056 11056 FALSE 23822 0.464108807
AdaptiveAvgPool1d bfloat16 [256 240 14] [256 240 1] noncontiguous fwd 46000 46000 FALSE 30489 1.508740857
AdaptiveAvgPool1d bfloat16 [256 240 14] [256 240 1] noncontiguous bwd 39696 79999 FALSE 23911 1.660156413
AdaptiveAvgPool1d bfloat16 [256 120 14] [256 120 1] contiguous fwd 18736 18736 FALSE 8889 2.107773653
AdaptiveAvgPool1d bfloat16 [256 120 14] [256 120 1] contiguous bwd 8784 8784 FALSE 13618 0.6450286386
AdaptiveAvgPool1d bfloat16 [256 120 14] [256 120 1] noncontiguous fwd 27552 27552 FALSE 22027 1.250828529
AdaptiveAvgPool1d bfloat16 [256 120 14] [256 120 1] noncontiguous bwd 26480 67392 FALSE 13547 1.954676312
AdaptiveAvgPool1d bfloat16 [256 144 14] [256 144 1] contiguous fwd 20352 20352 FALSE 9084 2.240422721
AdaptiveAvgPool1d bfloat16 [256 144 14] [256 144 1] contiguous bwd 9632 9632 FALSE 15698 0.6135813479
AdaptiveAvgPool1d bfloat16 [256 144 14] [256 144 1] noncontiguous fwd 28032 28032 FALSE 22062 1.270601033
AdaptiveAvgPool1d bfloat16 [256 144 14] [256 144 1] noncontiguous bwd 29184 64464 FALSE 15502 1.882595794
AdaptiveAvgPool1d bfloat16 [256 288 7] [256 288 1] contiguous fwd 20720 20720 FALSE 8142 2.54482928
AdaptiveAvgPool1d bfloat16 [256 288 7] [256 288 1] contiguous bwd 9600 9600 FALSE 15573 0.6164515508
AdaptiveAvgPool1d bfloat16 [256 288 7] [256 288 1] noncontiguous fwd 30736 30736 FALSE 19804 1.552009695
AdaptiveAvgPool1d bfloat16 [256 288 7] [256 288 1] noncontiguous bwd 29808 70191 FALSE 15716 1.896665818
AdaptiveAvgPool1d bfloat16 [256 576 7] [256 576 1] contiguous fwd 29728 29728 FALSE 10773 2.759491321
AdaptiveAvgPool1d bfloat16 [256 576 7] [256 576 1] contiguous bwd 12608 12608 FALSE 27858 0.4525809462
AdaptiveAvgPool1d bfloat16 [256 576 7] [256 576 1] noncontiguous fwd 44048 44048 FALSE 33226 1.325708782
AdaptiveAvgPool1d bfloat16 [256 576 7] [256 576 1] noncontiguous bwd 46080 87967 FALSE 28053 1.642605069
AdaptiveAvgPool1d bfloat16 [256 2048 7] [256 2048 1] contiguous fwd 75088 75088 FALSE 23680 3.170945946
AdaptiveAvgPool1d bfloat16 [256 2048 7] [256 2048 1] contiguous bwd 26256 26256 FALSE 92160 0.2848958333
AdaptiveAvgPool1d bfloat16 [256 2048 7] [256 2048 1] noncontiguous fwd 127664 127664 FALSE 97262 1.312578396
AdaptiveAvgPool1d bfloat16 [256 2048 7] [256 2048 1] noncontiguous bwd 145023 170207 FALSE 92284 1.571485848
AdaptiveAvgPool1d bfloat16 [1536 100 13] [1536 100 1] contiguous fwd 48720 48720 FALSE 24391 1.997458079
AdaptiveAvgPool1d bfloat16 [1536 100 13] [1536 100 1] contiguous bwd 16768 16768 FALSE 51235 0.327276276
AdaptiveAvgPool1d bfloat16 [1536 100 13] [1536 100 1] noncontiguous fwd 67024 67024 FALSE 53138 1.261319583
AdaptiveAvgPool1d bfloat16 [1536 100 13] [1536 100 1] noncontiguous bwd 73424 117759 FALSE 51520 1.42515528
AdaptiveAvgPool1d bfloat16 [731 100 13] [731 100 1] contiguous fwd 29760 29760 FALSE 13600 2.188235294
AdaptiveAvgPool1d bfloat16 [731 100 13] [731 100 1] contiguous bwd 11216 11216 FALSE 25849 0.4339045998
AdaptiveAvgPool1d bfloat16 [731 100 13] [731 100 1] noncontiguous fwd 29456 29456 FALSE 16444 1.791291657
AdaptiveAvgPool1d bfloat16 [731 100 13] [731 100 1] noncontiguous bwd 42592 85936 FALSE 26507 1.60682084
AdaptiveAvgPool1d bfloat16 [724 100 13] [724 100 1] contiguous fwd 29872 29872 FALSE 13422 2.225599762
AdaptiveAvgPool1d bfloat16 [724 100 13] [724 100 1] contiguous bwd 11552 11552 FALSE 25831 0.4472145871
AdaptiveAvgPool1d bfloat16 [724 100 13] [724 100 1] noncontiguous fwd 29120 29120 FALSE 16569 1.757498944
AdaptiveAvgPool1d bfloat16 [724 100 13] [724 100 1] noncontiguous bwd 42080 80976 FALSE 25884 1.625714727
AdaptiveAvgPool1d bfloat16 [64 16 80] [64 16 1] contiguous fwd 12896 12896 FALSE 23804 0.5417576878
AdaptiveAvgPool1d bfloat16 [64 16 80] [64 16 1] contiguous bwd 6144 6144 FALSE 4764 1.289672544
AdaptiveAvgPool1d bfloat16 [64 16 80] [64 16 1] noncontiguous fwd 36144 36144 FALSE 42417 0.8521111818
AdaptiveAvgPool1d bfloat16 [64 16 80] [64 16 1] noncontiguous bwd 16592 66112 FALSE 4782 3.469677959
AdaptiveAvgPool1d bfloat16 [64 96 20] [64 96 1] contiguous fwd 13200 13200 FALSE 9724 1.357466063
AdaptiveAvgPool1d bfloat16 [64 96 20] [64 96 1] contiguous bwd 7776 7776 FALSE 6116 1.271419228
AdaptiveAvgPool1d bfloat16 [64 96 20] [64 96 1] noncontiguous fwd 23264 23264 FALSE 15982 1.455637592
AdaptiveAvgPool1d bfloat16 [64 96 20] [64 96 1] noncontiguous bwd 17936 56591 FALSE 6382 2.810404262
AdaptiveAvgPool1d bfloat16 [64 240 20] [64 240 1] contiguous fwd 19104 19104 FALSE 9635 1.982771147
AdaptiveAvgPool1d bfloat16 [64 240 20] [64 240 1] contiguous bwd 8704 8704 FALSE 10613 0.8201262602
AdaptiveAvgPool1d bfloat16 [64 240 20] [64 240 1] noncontiguous fwd 25328 25328 FALSE 17244 1.468800742
AdaptiveAvgPool1d bfloat16 [64 240 20] [64 240 1] noncontiguous bwd 22192 62992 FALSE 11129 1.994069548
AdaptiveAvgPool1d bfloat16 [64 120 20] [64 120 1] contiguous fwd 14768 14768 FALSE 9760 1.513114754
AdaptiveAvgPool1d bfloat16 [64 120 20] [64 120 1] contiguous bwd 8208 8208 FALSE 6560 1.251219512
AdaptiveAvgPool1d bfloat16 [64 120 20] [64 120 1] noncontiguous fwd 23680 23680 FALSE 16480 1.436893204
AdaptiveAvgPool1d bfloat16 [64 120 20] [64 120 1] noncontiguous bwd 17408 62496 FALSE 6578 2.646397081
AdaptiveAvgPool1d bfloat16 [64 144 20] [64 144 1] contiguous fwd 13728 13728 FALSE 9600 1.43
AdaptiveAvgPool1d bfloat16 [64 144 20] [64 144 1] contiguous bwd 8064 8064 FALSE 7289 1.106324599
AdaptiveAvgPool1d bfloat16 [64 144 20] [64 144 1] noncontiguous fwd 23648 23648 FALSE 16320 1.449019608
AdaptiveAvgPool1d bfloat16 [64 144 20] [64 144 1] noncontiguous bwd 18816 92623 FALSE 7164 2.626465662
AdaptiveAvgPool1d bfloat16 [64 288 10] [64 288 1] contiguous fwd 13632 13632 FALSE 8036 1.696366351
AdaptiveAvgPool1d bfloat16 [64 288 10] [64 288 1] contiguous bwd 8384 8384 FALSE 7360 1.139130435
AdaptiveAvgPool1d bfloat16 [64 288 10] [64 288 1] noncontiguous fwd 18096 18096 FALSE 12249 1.477345089
AdaptiveAvgPool1d bfloat16 [64 288 10] [64 288 1] noncontiguous bwd 18560 92703 FALSE 7218 2.571349404
AdaptiveAvgPool1d bfloat16 [256 512 7] [256 512 7] contiguous fwd 231135 231135 FALSE 20533 11.25675741
AdaptiveAvgPool1d bfloat16 [256 512 7] [256 512 7] contiguous bwd 269663 293806 FALSE 25618 10.52630963
AdaptiveAvgPool1d bfloat16 [256 512 7] [256 512 7] noncontiguous fwd 261663 278142 FALSE 31342 8.348637611
AdaptiveAvgPool1d bfloat16 [256 512 7] [256 512 7] noncontiguous bwd 300798 336483 FALSE 31483 9.554299146
AdaptiveAvgPool 2D
op_name dtype input_size output_size contiguous direction rocm_kernel_avg rocm_op_avg use_miopen MIOPEN MIOPEN_over_Rocm
AdaptiveAvgPool2d float16 [256 256 6 6] [256 256 6 6] contiguous fwd 298910 298910 FALSE 111839 2.672681265
AdaptiveAvgPool2d float16 [256 256 6 6] [256 256 6 6] contiguous bwd 303534 322878 FALSE 179181 1.694007735
AdaptiveAvgPool2d float16 [256 256 6 6] [256 256 6 6] noncontiguous fwd 364655 381151 FALSE 114044 3.19749395
AdaptiveAvgPool2d float16 [256 256 6 6] [256 256 6 6] noncontiguous bwd 373054 409534 FALSE 181226 2.05850154
AdaptiveAvgPool2d float16 [16 72 64 64] [16 72 1 1] contiguous fwd 19152 19152 FALSE 1818587 0.01053125311
AdaptiveAvgPool2d float16 [16 72 64 64] [16 72 1 1] contiguous bwd 26176 26176 FALSE 354060 0.07393097215
AdaptiveAvgPool2d float16 [16 72 64 64] [16 72 1 1] noncontiguous fwd 79040 79040 FALSE 1712259 0.04616124079
AdaptiveAvgPool2d float16 [16 72 64 64] [16 72 1 1] noncontiguous bwd 190319 222319 FALSE 353989 0.5376410001
AdaptiveAvgPool2d float16 [16 120 64 64] [16 120 1 1] contiguous fwd 22784 22784 FALSE 1820453 0.01251556618
AdaptiveAvgPool2d float16 [16 120 64 64] [16 120 1 1] contiguous bwd 40816 40816 FALSE 587162 0.06951403531
AdaptiveAvgPool2d float16 [16 120 64 64] [16 120 1 1] noncontiguous fwd 127920 127920 FALSE 1799920 0.07106982533
AdaptiveAvgPool2d float16 [16 120 64 64] [16 120 1 1] noncontiguous bwd 1206059 1229611 FALSE 587019 2.05454849
AdaptiveAvgPool2d float16 [64 672 13 17] [64 672 1 1] noncontiguous fwd 1111499 1111499 FALSE 254417 4.368807902
AdaptiveAvgPool2d float16 [256 2048 7 7] [256 2048 1 1] noncontiguous fwd 3985614 3985614 FALSE 851925 4.678362532
AdaptiveAvgPool2d float32 [47 120 25 38] [47 120 1 1] noncontiguous bwd 420366 438046 FALSE 401420 1.047197449
AdaptiveAvgPool2d float32 [54 480 13 19] [54 480 1 1] contiguous fwd 34832 34832 FALSE 113759 0.3061911585
AdaptiveAvgPool2d float32 [54 480 13 19] [54 480 1 1] contiguous bwd 38480 38480 FALSE 478700 0.08038437435
AdaptiveAvgPool2d float32 [54 480 13 19] [54 480 1 1] noncontiguous fwd 618141 618141 FALSE 175839 3.515380547
AdaptiveAvgPool2d float32 [54 480 13 19] [54 480 1 1] noncontiguous bwd 605405 622749 FALSE 478540 1.265108455
AdaptiveAvgPool2d float32 [47 480 13 19] [47 480 1 1] contiguous fwd 32256 32256 FALSE 113546 0.2840786994
AdaptiveAvgPool2d float32 [47 480 13 19] [47 480 1 1] contiguous bwd 34288 34288 FALSE 417385 0.08214957413
AdaptiveAvgPool2d float32 [47 480 13 19] [47 480 1 1] noncontiguous fwd 2344662 2344662 FALSE 156639 14.96857105
AdaptiveAvgPool2d float32 [47 480 13 19] [47 480 1 1] noncontiguous bwd 494846 512526 FALSE 417420 1.185487039
AdaptiveAvgPool2d float32 [54 672 13 19] [54 672 1 1] contiguous fwd 43632 43632 FALSE 189795 0.2298901446
AdaptiveAvgPool2d float32 [54 672 13 19] [54 672 1 1] contiguous bwd 52832 52832 FALSE 668637 0.07901447273
AdaptiveAvgPool2d float32 [54 672 13 19] [54 672 1 1] noncontiguous fwd 974300 974300 FALSE 262985 3.704774037
AdaptiveAvgPool2d float32 [54 672 13 19] [54 672 1 1] noncontiguous bwd 877820 898604 FALSE 668690 1.312745816
AdaptiveAvgPool2d float32 [47 672 13 19] [47 672 1 1] contiguous fwd 39616 39616 FALSE 180070 0.220003332
AdaptiveAvgPool2d float32 [47 672 13 19] [47 672 1 1] contiguous bwd 45584 45584 FALSE 583428 0.0781313204
AdaptiveAvgPool2d float32 [47 672 13 19] [47 672 1 1] noncontiguous fwd 807612 807612 FALSE 255998 3.154759022
AdaptiveAvgPool2d float32 [47 672 13 19] [47 672 1 1] noncontiguous bwd 788204 806236 FALSE 583055 1.351851884
AdaptiveAvgPool2d float32 [47 960 13 19] [47 960 1 1] contiguous fwd 52432 52432 FALSE 263234 0.199183996
AdaptiveAvgPool2d float32 [47 960 13 19] [47 960 1 1] contiguous bwd 64112 64112 FALSE 830431 0.0772032836
AdaptiveAvgPool2d float32 [47 960 13 19] [47 960 1 1] noncontiguous fwd 1261722 1261722 FALSE 297314 4.243735579
AdaptiveAvgPool2d float32 [47 960 13 19] [47 960 1 1] noncontiguous bwd 1172715 1190155 FALSE 830698 1.411722431
AdaptiveAvgPool2d float32 [54 960 13 19] [54 960 1 1] contiguous fwd 57520 57520 FALSE 330185 0.1742053697
AdaptiveAvgPool2d float32 [54 960 13 19] [54 960 1 1] contiguous bwd 73408 73408 FALSE 953222 0.07701039212
AdaptiveAvgPool2d float32 [54 960 13 19] [54 960 1 1] noncontiguous fwd 1513737 1513737 FALSE 336763 4.494962333
AdaptiveAvgPool2d float32 [54 960 13 19] [54 960 1 1] noncontiguous bwd 1363786 1382330 FALSE 953364 1.430498739
AdaptiveAvgPool2d bfloat16 [64 288 10 10] [64 288 1 1] contiguous fwd 52224 52224 FALSE 30862 1.692178083
AdaptiveAvgPool2d bfloat16 [64 288 10 10] [64 288 1 1] contiguous bwd 16176 16176 FALSE 141759 0.1141091571
AdaptiveAvgPool2d bfloat16 [64 288 10 10] [64 288 1 1] noncontiguous fwd 101168 101168 FALSE 76693 1.319129516
AdaptiveAvgPool2d bfloat16 [64 288 10 10] [64 288 1 1] noncontiguous bwd 71824 117536 FALSE 141724 0.5067878412
AdaptiveAvgPool2d bfloat16 [256 512 7 7] [256 512 7 7] contiguous fwd 766220 766220 FALSE 301385 2.542329578
AdaptiveAvgPool2d bfloat16 [256 512 7 7] [256 512 7 7] contiguous bwd 884892 920252 FALSE 485278 1.82347438
AdaptiveAvgPool2d bfloat16 [256 512 7 7] [256 512 7 7] noncontiguous fwd 933820 949468 FALSE 308443 3.027528587
AdaptiveAvgPool2d bfloat16 [256 512 7 7] [256 512 7 7] noncontiguous bwd 1136267 1169211 FALSE 490985 2.31426011
AdaptiveAvgPool 3D
op_name dtype input_size output_size contiguous direction rocm_kernel_avg rocm_op_avg use_miopen Miopen Miopen_over_Rocm
AdaptiveAvgPool3d float16 [64 512 2 7 7] [64 512 1 1 1] contiguous fwd 79440 79440 FALSE 74987 1.059383626
AdaptiveAvgPool3d float16 [64 512 2 7 7] [64 512 1 1 1] contiguous bwd 21312 21312 FALSE 182630 0.116694957
AdaptiveAvgPool3d float16 [64 512 2 7 7] [64 512 1 1 1] noncontiguous fwd 240223 240223 FALSE 111768 2.149300336
AdaptiveAvgPool3d float16 [128 512 2 7 7] [128 512 1 1 1] noncontiguous fwd 611581 611581 FALSE 192888 3.170653436
AdaptiveAvgPool3d float16 [724 100 13 13 13] [724 100 1 1 1] noncontiguous fwd 11422890 11422890 FALSE 9138836 1.249928328
AdaptiveAvgPool3d float16 [724 100 13 13 13] [724 100 1 1 1] noncontiguous bwd 13308401 13325665 FALSE 8931583 1.490038328
AdaptiveAvgPool3d float32 [47 480 13 19 19] [47 480 1 1 1] noncontiguous fwd 6979311 6979311 FALSE 5141542 1.357435376
AdaptiveAvgPool3d float32 [47 480 13 19 19] [47 480 1 1 1] noncontiguous bwd 11083324 11100744 FALSE 5925397 1.870477877
AdaptiveAvgPool3d float32 [54 672 13 19 19] [54 672 1 1 1] noncontiguous fwd 12506677 12506677 FALSE 9639852 1.297393051
AdaptiveAvgPool3d float32 [54 672 13 19 19] [54 672 1 1 1] noncontiguous bwd 19934802 19952050 FALSE 9554856 2.08635295
AdaptiveAvgPool3d float32 [47 672 13 19 19] [47 672 1 1 1] noncontiguous fwd 10268991 10268991 FALSE 9070822 1.132090454
AdaptiveAvgPool3d float32 [47 672 13 19 19] [47 672 1 1 1] noncontiguous bwd 16737745 16755025 FALSE 8317705 2.012303274
AdaptiveAvgPool3d float32 [64 96 20 20 20] [64 96 1 1 1] noncontiguous bwd 5618950 5636230 FALSE 2748380 2.044458918
AdaptiveAvgPool3d float32 [64 240 20 20 20] [64 240 1 1 1] contiguous fwd 376542 376542 FALSE 4202934 0.08959027194
AdaptiveAvgPool3d float32 [64 240 20 20 20] [64 240 1 1 1] contiguous bwd 613597 613597 FALSE 6874799 0.08925308216
AdaptiveAvgPool3d float32 [64 240 20 20 20] [64 240 1 1 1] noncontiguous fwd 10019553 10019553 FALSE 12288268 0.8153755273
AdaptiveAvgPool3d float32 [64 240 20 20 20] [64 240 1 1 1] noncontiguous bwd 14018878 14036158 FALSE 6874016 2.039401421
AdaptiveAvgPool3d float32 [64 120 20 20 20] [64 120 1 1 1] contiguous fwd 200943 200943 FALSE 3999166 0.05024622634
AdaptiveAvgPool3d float32 [64 120 20 20 20] [64 120 1 1 1] contiguous bwd 309390 309390 FALSE 3434564 0.0900813029
AdaptiveAvgPool3d float32 [64 120 20 20 20] [64 120 1 1 1] noncontiguous fwd 4574634 4574634 FALSE 8648018 0.5289806289
AdaptiveAvgPool3d float32 [64 120 20 20 20] [64 120 1 1 1] noncontiguous bwd 7029615 7046943 FALSE 3434758 2.046611435
AdaptiveAvgPool3d float32 [64 144 20 20 20] [64 144 1 1 1] contiguous fwd 233807 233807 FALSE 3993371 0.05854877997
AdaptiveAvgPool3d float32 [64 144 20 20 20] [64 144 1 1 1] contiguous bwd 370766 370766 FALSE 4122347 0.08994051204
AdaptiveAvgPool3d float32 [64 144 20 20 20] [64 144 1 1 1] noncontiguous fwd 5731461 5731461 FALSE 11300220 0.5071990634
AdaptiveAvgPool3d float32 [64 144 20 20 20] [64 144 1 1 1] noncontiguous bwd 8423976 8441192 FALSE 4121583 2.043869067
AdaptiveAvgPool3d float32 [64 288 10 10 10] [64 288 1 1 1] contiguous fwd 65936 65936 FALSE 234470 0.2812129484
AdaptiveAvgPool3d float32 [64 288 10 10 10] [64 288 1 1 1] contiguous bwd 98864 98864 FALSE 2575136 0.03839175873
AdaptiveAvgPool3d float32 [64 288 10 10 10] [64 288 1 1 1] noncontiguous fwd 1758472 1758472 FALSE 860796 2.042844065
AdaptiveAvgPool3d float32 [64 288 10 10 10] [64 288 1 1 1] noncontiguous bwd 1812391 1829655 FALSE 2571048 0.7049230508
AdaptiveAvgPool3d float32 [256 512 7 7 7] [256 512 7 7 7] contiguous fwd 7120494 7346109 FALSE 1706624 4.17226876
AdaptiveAvgPool3d float32 [256 512 7 7 7] [256 512 7 7 7] contiguous bwd 11550874 11794088 FALSE 2523297 4.577691013
AdaptiveAvgPool3d float32 [256 512 7 7 7] [256 512 7 7 7] noncontiguous fwd 11461178 11703689 FALSE 5492901 2.086543704
AdaptiveAvgPool3d float32 [256 512 7 7 7] [256 512 7 7 7] noncontiguous bwd 15649686 15907061 FALSE 5483640 2.85388647
AdaptiveAvgPool3d bfloat16 [64 512 2 7 7] [64 512 1 1 1] contiguous fwd 80624 80624 FALSE 75342 1.070106979
AdaptiveAvgPool3d bfloat16 [64 512 2 7 7] [64 512 1 1 1] contiguous bwd 22960 22960 FALSE 184337 0.1245544844
AdaptiveAvgPool3d bfloat16 [64 512 2 7 7] [64 512 1 1 1] noncontiguous fwd 236735 236735 FALSE 111964 2.114384981
AdaptiveAvgPool3d bfloat16 [256 1280 7 7 7] [256 1280 1 1 1] noncontiguous fwd 15784134 15784134 FALSE 4663839 3.384365112
AdaptiveAvgPool3d bfloat16 [256 1280 7 7 7] [256 1280 1 1 1] noncontiguous bwd 9144197 9161605 FALSE 6370996 1.435285315
AdaptiveAvgPool3d bfloat16 [256 16 56 56 56] [256 16 1 1 1] noncontiguous bwd 78386130 78404858 FALSE 40826286 1.919991694
AdaptiveAvgPool3d bfloat16 [256 96 14 14 14] [256 96 1 1 1] noncontiguous fwd 5011993 5011993 FALSE 2433234 2.059807236
AdaptiveAvgPool3d bfloat16 [256 96 14 14 14] [256 96 1 1 1] noncontiguous bwd 6056116 6073508 FALSE 3820056 1.58534744
AdaptiveAvgPool3d bfloat16 [256 240 14 14 14] [256 240 1 1 1] noncontiguous fwd 15253193 15253193 FALSE 12100505 1.260541853
AdaptiveAvgPool3d bfloat16 [256 240 14 14 14] [256 240 1 1 1] noncontiguous bwd 15618327 15635719 FALSE 9573633 1.631389776
AdaptiveAvgPool3d bfloat16 [256 120 14 14 14] [256 120 1 1 1] noncontiguous fwd 6772416 6772416 FALSE 6105505 1.109231096
AdaptiveAvgPool3d bfloat16 [256 120 14 14 14] [256 120 1 1 1] noncontiguous bwd 7442653 7461675 FALSE 4774202 1.558931315
AdaptiveAvgPool3d bfloat16 [256 144 14 14 14] [256 144 1 1 1] noncontiguous fwd 8442153 8442153 FALSE 6510534 1.296691331
AdaptiveAvgPool3d bfloat16 [256 144 14 14 14] [256 144 1 1 1] noncontiguous bwd 9173765 9191691 FALSE 5730856 1.600766971
AdaptiveAvgPool3d bfloat16 [256 288 7 7 7] [256 288 1 1 1] noncontiguous fwd 2235862 2235862 FALSE 750753 2.978159261
AdaptiveAvgPool3d bfloat16 [256 576 7 7 7] [256 576 1 1 1] noncontiguous fwd 6306130 6306130 FALSE 1461434 4.315028937
AdaptiveAvgPool3d bfloat16 [256 576 7 7 7] [256 576 1 1 1] noncontiguous bwd 3398400 3415840 FALSE 2864380 1.186434761
AdaptiveAvgPool3d bfloat16 [256 2048 7 7 7] [256 2048 1 1 1] noncontiguous fwd 25332217 25332217 FALSE 8133265 3.114643012
AdaptiveAvgPool3d bfloat16 [256 2048 7 7 7] [256 2048 1 1 1] noncontiguous bwd 16092516 16110020 FALSE 10197861 1.578028569
AdaptiveAvgPool3d bfloat16 [1536 100 13 13 13] [1536 100 1 1 1] noncontiguous fwd 26991425 26991425 FALSE 22092018 1.221772724
AdaptiveAvgPool3d bfloat16 [1536 100 13 13 13] [1536 100 1 1 1] noncontiguous bwd 28571274 28588778 FALSE 19160334 1.491167847
AdaptiveAvgPool3d bfloat16 [731 100 13 13 13] [731 100 1 1 1] noncontiguous fwd 11556890 11556890 FALSE 9150578 1.262968306
AdaptiveAvgPool3d bfloat16 [731 100 13 13 13] [731 100 1 1 1] noncontiguous bwd 13468081 13485569 FALSE 9112266 1.478016665
AdaptiveAvgPool3d bfloat16 [64 144 20 20 20] [64 144 1 1 1] noncontiguous bwd 6803664 6821824 FALSE 4174719 1.629729809
AdaptiveAvgPool3d bfloat16 [64 288 10 10 10] [64 288 1 1 1] noncontiguous fwd 1319914 1319914 FALSE 858948 1.536663453
AdaptiveAvgPool3d bfloat16 [64 288 10 10 10] [64 288 1 1 1] noncontiguous bwd 1217498 1234842 FALSE 1044778 1.165317417
AdaptiveAvgPool3d bfloat16 [256 512 7 7 7] [256 512 7 7 7] contiguous fwd 7252109 7480109 FALSE 1732953 4.184827286
AdaptiveAvgPool3d bfloat16 [256 512 7 7 7] [256 512 7 7 7] contiguous bwd 11783368 12028568 FALSE 2547954 4.62463922
AdaptiveAvgPool3d bfloat16 [256 512 7 7 7] [256 512 7 7 7] noncontiguous fwd 11162716 11407067 FALSE 4997260 2.233767304
AdaptiveAvgPool3d bfloat16 [256 512 7 7 7] [256 512 7 7 7] noncontiguous bwd 14863981 15134953 FALSE 5004957 2.969851889

hieule88 added 29 commits August 5, 2024 18:11
@hieule88
init
d197108
@hieule88
skeleton code
3c90908
@hieule88
add solver, kernel
86a5073
@hieule88
add gtest
ca4ad97
@hieule88
add driver test
0492fc7
@hieule88
change kinfor to ksize
881e796
@hieule88
change params
3612897
@hieule88
fix gtest
930d47e
@hieule88
passed gtest
5a35738
@hieule88
skeleton code
27470a2
@hieule88
small fix
2e131c7
@hieule88
Merge remote-tracking branch 'origin/develop-moreh' into impl_adaptiv…
0e773fa 
…e_avgpool
@hieule88
add isOverRocm3d
7d3a0a6
@hieule88
add const Tgpu
dfbb6c7
@hieule88
add resolve comment Rocm PR
88179b0
@hieule88
add issametype
d2b2d1f
@hieule88
rm ENV gtest
0dc61de
@hieule88
rm magic number in driver test
194ab40
@hieule88
change to AddTensorFlag
36b0662
@hieule88
small fix
99ff4bb
@hieule88
fix driver as comments
42b54a7
@hieule88
for to parford
22ae98a
@hieule88
resolved comments
6ef45f6
@hieule88
rm large number of cast in kernel
796adb1
@hieule88
resolved comments
93985d4
@hieule88
add T
7795a19
@hieule88
Merge remote-tracking branch 'upstream/develop' into impl_adaptive_av…
cde0f8b 
…gpool
@hieule88
Merge remote-tracking branch 'upstream/develop' into impl_adaptive_av…
eabac2a 
…gpool
@hieule88
reorder
58bcf1e
@hieule88 hieule88 self-assigned this Oct 18, 2024
@hieule88 hieule88 force-pushed the impl_adaptive_avgpool branch from ae71d69 to bd72a3e Compare October 18, 2024 07:03
@iq136boy
Copy link
Contributor

CI passed. But the PR still need to solve conflict before it is getting merged.

iq136boy
iq136boy reviewed Nov 19, 2024
View reviewed changes
Copy link
Contributor

@iq136boy iq136boy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Error_log:
3301_log.txt

@hieule88
Copy link
Collaborator Author

Error_log: 3301_log.txt

I fixed this error. Please send me the newest error log of this PR.

iq136boy
iq136boy reviewed Dec 4, 2024
View reviewed changes
Copy link
Contributor

@iq136boy iq136boy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

error log:
3301_log (2).txt

iq136boy
iq136boy reviewed Jan 14, 2025
View reviewed changes
Copy link
Contributor

@iq136boy iq136boy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

error_log:
3301_log3.txt

@long10024070
Copy link
Collaborator

Because we have other pooling operations nearly similar to AdaptiveAvgpool, this PR should be reviewed soon. Therefore, I added "urgency_normal" label to this PR.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@iq136boy iq136boy iq136boy left review comments

@JehandadKhan JehandadKhan Awaiting requested review from JehandadKhan

@junliume junliume Awaiting requested review from junliume

@BrianHarrisonAMD BrianHarrisonAMD Awaiting requested review from BrianHarrisonAMD BrianHarrisonAMD is a code owner

@BradPepersAMD BradPepersAMD Awaiting requested review from BradPepersAMD BradPepersAMD is a code owner

At least 1 approving review is required to merge this pull request.

Assignees

@hieule88 hieule88

Labels
enhancement external_collaborator urgency_normal
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@hieule88 @iq136boy @long10024070