pytorch 2.2.1

AMD Ryzen Threadripper 7980X 64-Cores testing with a System76 Thelio Major (FA Z5 BIOS) and AMD Radeon Pro W7900 45GB on Ubuntu 23.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2403260-PTS-PYTORCH257&rdt&grr.

pytorch 2.2.1ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionabcdAMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads)System76 Thelio Major (FA Z5 BIOS)AMD Device 14a44 x 32GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA21000GB CT1000T700SSD5AMD Radeon Pro W7900 45GB (1760/1124MHz)AMD Device 14ccDELL P2415QAquantia AQC113C NBase-T/IEEE + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 23.106.5.0-26-generic (x86_64)GNOME Shell 45.2X Server + Wayland4.6 Mesa 23.2.1-1ubuntu3.1 (LLVM 15.0.7 DRM 3.54)GCC 13.2.0ext41920x1080OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseProcessor Details- Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105Python Details- Python 3.11.6Security Details- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

pytorch 2.2.1pytorch: CPU - 64 - Efficientnet_v2_lpytorch: CPU - 256 - Efficientnet_v2_lpytorch: CPU - 16 - Efficientnet_v2_lpytorch: CPU - 32 - Efficientnet_v2_lpytorch: CPU - 512 - Efficientnet_v2_lpytorch: CPU - 512 - ResNet-152pytorch: CPU - 16 - ResNet-152pytorch: CPU - 256 - ResNet-152pytorch: CPU - 32 - ResNet-152pytorch: CPU - 64 - ResNet-152pytorch: CPU - 1 - Efficientnet_v2_lpytorch: CPU - 1 - ResNet-152pytorch: CPU - 32 - ResNet-50pytorch: CPU - 512 - ResNet-50pytorch: CPU - 16 - ResNet-50pytorch: CPU - 64 - ResNet-50pytorch: CPU - 256 - ResNet-50pytorch: CPU - 1 - ResNet-50abcd7.787.787.807.817.7818.8118.7119.0219.0018.9711.9921.8248.4347.8448.1048.2548.5059.877.807.777.817.797.7919.0519.4319.4718.9319.3411.9421.8746.9148.1248.0447.8447.9860.447.767.747.827.777.7918.7419.1818.8819.0918.5112.0021.7547.3548.1347.5947.7947.6359.767.807.817.747.797.7519.1618.6218.9219.3719.4312.0522.0447.1147.2747.9047.5248.7159.78OpenBenchmarking.org

PyTorch

Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_labcd246810SE +/- 0.02, N = 37.787.807.767.80MIN: 7.07 / MAX: 8.15MIN: 7.04 / MAX: 8.09MIN: 7.17 / MAX: 8.13MIN: 7.12 / MAX: 8.13

PyTorch

Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_labcd246810SE +/- 0.01, N = 37.787.777.747.81MIN: 7.1 / MAX: 8.12MIN: 7.09 / MAX: 8.11MIN: 7 / MAX: 8.06MIN: 7.08 / MAX: 8.17

PyTorch

Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_labcd246810SE +/- 0.02, N = 37.807.817.827.74MIN: 7.02 / MAX: 8.19MIN: 7.14 / MAX: 8.1MIN: 7.22 / MAX: 8.18MIN: 7 / MAX: 8.01

PyTorch

Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_labcd246810SE +/- 0.01, N = 37.817.797.777.79MIN: 7.09 / MAX: 8.15MIN: 7.08 / MAX: 8.09MIN: 7.04 / MAX: 8.06MIN: 7.2 / MAX: 8.08

PyTorch

Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_labcd246810SE +/- 0.02, N = 37.787.797.797.75MIN: 7.11 / MAX: 8.12MIN: 7.21 / MAX: 8.14MIN: 7.14 / MAX: 8.11MIN: 7.2 / MAX: 8.05

PyTorch

Device: CPU - Batch Size: 512 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-152abcd510152025SE +/- 0.12, N = 318.8119.0518.7419.16MIN: 18.12 / MAX: 19.1MIN: 18.53 / MAX: 19.21MIN: 18.24 / MAX: 18.88MIN: 18.7 / MAX: 19.35

PyTorch

Device: CPU - Batch Size: 16 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-152abcd510152025SE +/- 0.07, N = 318.7119.4319.1818.62MIN: 18.06 / MAX: 18.98MIN: 18.98 / MAX: 19.61MIN: 18.6 / MAX: 19.36MIN: 18.1 / MAX: 18.77

PyTorch

Device: CPU - Batch Size: 256 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-152abcd510152025SE +/- 0.05, N = 319.0219.4718.8818.92MIN: 18.43 / MAX: 19.26MIN: 18.9 / MAX: 19.63MIN: 18.14 / MAX: 19.04MIN: 18.19 / MAX: 19.1

PyTorch

Device: CPU - Batch Size: 32 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-152abcd510152025SE +/- 0.02, N = 319.0018.9319.0919.37MIN: 18.39 / MAX: 19.17MIN: 18.35 / MAX: 19.07MIN: 18.53 / MAX: 19.25MIN: 18.65 / MAX: 19.53

PyTorch

Device: CPU - Batch Size: 64 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-152abcd510152025SE +/- 0.15, N = 318.9719.3418.5119.43MIN: 18.22 / MAX: 19.28MIN: 18.65 / MAX: 19.48MIN: 17.93 / MAX: 18.68MIN: 19.02 / MAX: 19.6

PyTorch

Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_labcd3691215SE +/- 0.09, N = 311.9911.9412.0012.05MIN: 11.68 / MAX: 12.24MIN: 11.71 / MAX: 12.13MIN: 10.62 / MAX: 12.16MIN: 11.9 / MAX: 12.2

PyTorch

Device: CPU - Batch Size: 1 - Model: ResNet-152

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-152abcd510152025SE +/- 0.08, N = 321.8221.8721.7522.04MIN: 21.05 / MAX: 22.15MIN: 20.98 / MAX: 22.14MIN: 20.95 / MAX: 21.96MIN: 21.04 / MAX: 22.24

PyTorch

Device: CPU - Batch Size: 32 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-50abcd1122334455SE +/- 0.09, N = 348.4346.9147.3547.11MIN: 43.88 / MAX: 49.3MIN: 43.81 / MAX: 47.55MIN: 43.64 / MAX: 48.24MIN: 44.02 / MAX: 47.74

PyTorch

Device: CPU - Batch Size: 512 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-50abcd1122334455SE +/- 0.10, N = 347.8448.1248.1347.27MIN: 43.86 / MAX: 48.56MIN: 44.65 / MAX: 48.96MIN: 43.45 / MAX: 48.9MIN: 43.83 / MAX: 48.07

PyTorch

Device: CPU - Batch Size: 16 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-50abcd1122334455SE +/- 0.25, N = 348.1048.0447.5947.90MIN: 42 / MAX: 49.24MIN: 44.19 / MAX: 48.74MIN: 44.44 / MAX: 48.39MIN: 44.12 / MAX: 48.66

PyTorch

Device: CPU - Batch Size: 64 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-50abcd1122334455SE +/- 0.16, N = 348.2547.8447.7947.52MIN: 43.55 / MAX: 49.28MIN: 44.44 / MAX: 48.63MIN: 44.49 / MAX: 48.52MIN: 43.99 / MAX: 48.23

PyTorch

Device: CPU - Batch Size: 256 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-50abcd1122334455SE +/- 0.17, N = 348.5047.9847.6348.71MIN: 44.96 / MAX: 49.46MIN: 43.06 / MAX: 48.63MIN: 43.91 / MAX: 48.42MIN: 44.82 / MAX: 49.49

PyTorch

Device: CPU - Batch Size: 1 - Model: ResNet-50

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-50abcd1428425670SE +/- 0.10, N = 359.8760.4459.7659.78MIN: 53.95 / MAX: 61.32MIN: 55.61 / MAX: 61.31MIN: 54.51 / MAX: 60.81MIN: 53.76 / MAX: 60.69


Phoronix Test Suite v10.8.5