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.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2403260-PTS-PYTORCH257
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
a
March 26
  1 Hour, 47 Minutes
b
March 26
  36 Minutes
c
March 26
  36 Minutes
d
March 26
  36 Minutes
Invert Hiding All Results Option
  54 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

a

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.11.6
Security Notes: 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

b

c

d

Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads), Motherboard: System76 Thelio Major (FA Z5 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA2, Disk: 1000GB CT1000T700SSD5, Graphics: AMD Radeon Pro W7900 45GB (1760/1124MHz), Audio: AMD Device 14cc, Monitor: DELL P2415Q, Network: Aquantia AQC113C NBase-T/IEEE + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.10, Kernel: 6.5.0-26-generic (x86_64), Desktop: GNOME Shell 45.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 23.2.1-1ubuntu3.1 (LLVM 15.0.7 DRM 3.54), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080

pytorch 2.2.1OpenBenchmarking.orgPhoronix Test SuiteAMD 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 13.2.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionPytorch 2.2.1 BenchmarksSystem Logs- Transparent Huge Pages: madvise- Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105- Python 3.11.6- 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

abcdResult OverviewPhoronix Test Suite100%101%103%104%105%PyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchCPU - 64 - ResNet-152CPU - 16 - ResNet-152CPU - 32 - ResNet-50CPU - 256 - ResNet-152CPU - 32 - ResNet-152CPU - 256 - ResNet-50CPU - 512 - ResNet-152CPU - 512 - ResNet-50CPU - 64 - ResNet-50CPU - 1 - ResNet-152CPU - 1 - ResNet-50CPU - 16 - ResNet-50CPU - 16 - Efficientnet_v2_lCPU - 1 - Efficientnet_v2_lCPU - 256 - Efficientnet_v2_lCPU - 512 - Efficientnet_v2_lCPU - 64 - Efficientnet_v2_lCPU - 32 - Efficientnet_v2_l

pytorch 2.2.1pytorch: CPU - 1 - ResNet-50pytorch: CPU - 1 - ResNet-152pytorch: CPU - 16 - ResNet-50pytorch: CPU - 32 - ResNet-50pytorch: CPU - 64 - ResNet-50pytorch: CPU - 16 - ResNet-152pytorch: CPU - 256 - ResNet-50pytorch: CPU - 32 - ResNet-152pytorch: CPU - 512 - ResNet-50pytorch: CPU - 64 - ResNet-152pytorch: CPU - 256 - ResNet-152pytorch: CPU - 512 - ResNet-152pytorch: CPU - 1 - Efficientnet_v2_lpytorch: CPU - 16 - Efficientnet_v2_lpytorch: CPU - 32 - Efficientnet_v2_lpytorch: CPU - 64 - Efficientnet_v2_lpytorch: CPU - 256 - Efficientnet_v2_lpytorch: CPU - 512 - Efficientnet_v2_labcd59.8721.8248.1048.4348.2518.7148.5019.0047.8418.9719.0218.8111.997.807.817.787.787.7860.4421.8748.0446.9147.8419.4347.9818.9348.1219.3419.4719.0511.947.817.797.807.777.7959.7621.7547.5947.3547.7919.1847.6319.0948.1318.5118.8818.7412.007.827.777.767.747.7959.7822.0447.9047.1147.5218.6248.7119.3747.2719.4318.9219.1612.057.747.797.807.817.75OpenBenchmarking.org

PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-50cdab1428425670SE +/- 0.10, N = 359.7659.7859.8760.44MIN: 54.51 / MAX: 60.81MIN: 53.76 / MAX: 60.69MIN: 53.95 / MAX: 61.32MIN: 55.61 / MAX: 61.31

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-152cabd510152025SE +/- 0.08, N = 321.7521.8221.8722.04MIN: 20.95 / MAX: 21.96MIN: 21.05 / MAX: 22.15MIN: 20.98 / MAX: 22.14MIN: 21.04 / MAX: 22.24

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-50cdba1122334455SE +/- 0.25, N = 347.5947.9048.0448.10MIN: 44.44 / MAX: 48.39MIN: 44.12 / MAX: 48.66MIN: 44.19 / MAX: 48.74MIN: 42 / MAX: 49.24

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-50bdca1122334455SE +/- 0.09, N = 346.9147.1147.3548.43MIN: 43.81 / MAX: 47.55MIN: 44.02 / MAX: 47.74MIN: 43.64 / MAX: 48.24MIN: 43.88 / MAX: 49.3

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-50dcba1122334455SE +/- 0.16, N = 347.5247.7947.8448.25MIN: 43.99 / MAX: 48.23MIN: 44.49 / MAX: 48.52MIN: 44.44 / MAX: 48.63MIN: 43.55 / MAX: 49.28

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-152dacb510152025SE +/- 0.07, N = 318.6218.7119.1819.43MIN: 18.1 / MAX: 18.77MIN: 18.06 / MAX: 18.98MIN: 18.6 / MAX: 19.36MIN: 18.98 / MAX: 19.61

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-50cbad1122334455SE +/- 0.17, N = 347.6347.9848.5048.71MIN: 43.91 / MAX: 48.42MIN: 43.06 / MAX: 48.63MIN: 44.96 / MAX: 49.46MIN: 44.82 / MAX: 49.49

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-152bacd510152025SE +/- 0.02, N = 318.9319.0019.0919.37MIN: 18.35 / MAX: 19.07MIN: 18.39 / MAX: 19.17MIN: 18.53 / MAX: 19.25MIN: 18.65 / MAX: 19.53

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-50dabc1122334455SE +/- 0.10, N = 347.2747.8448.1248.13MIN: 43.83 / MAX: 48.07MIN: 43.86 / MAX: 48.56MIN: 44.65 / MAX: 48.96MIN: 43.45 / MAX: 48.9

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-152cabd510152025SE +/- 0.15, N = 318.5118.9719.3419.43MIN: 17.93 / MAX: 18.68MIN: 18.22 / MAX: 19.28MIN: 18.65 / MAX: 19.48MIN: 19.02 / MAX: 19.6

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-152cdab510152025SE +/- 0.05, N = 318.8818.9219.0219.47MIN: 18.14 / MAX: 19.04MIN: 18.19 / MAX: 19.1MIN: 18.43 / MAX: 19.26MIN: 18.9 / MAX: 19.63

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-152cabd510152025SE +/- 0.12, N = 318.7418.8119.0519.16MIN: 18.24 / MAX: 18.88MIN: 18.12 / MAX: 19.1MIN: 18.53 / MAX: 19.21MIN: 18.7 / MAX: 19.35

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_lbacd3691215SE +/- 0.09, N = 311.9411.9912.0012.05MIN: 11.71 / MAX: 12.13MIN: 11.68 / MAX: 12.24MIN: 10.62 / MAX: 12.16MIN: 11.9 / MAX: 12.2

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_ldabc246810SE +/- 0.02, N = 37.747.807.817.82MIN: 7 / MAX: 8.01MIN: 7.02 / MAX: 8.19MIN: 7.14 / MAX: 8.1MIN: 7.22 / MAX: 8.18

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_lcbda246810SE +/- 0.01, N = 37.777.797.797.81MIN: 7.04 / MAX: 8.06MIN: 7.08 / MAX: 8.09MIN: 7.2 / MAX: 8.08MIN: 7.09 / MAX: 8.15

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_lcabd246810SE +/- 0.02, N = 37.767.787.807.80MIN: 7.17 / MAX: 8.13MIN: 7.07 / MAX: 8.15MIN: 7.04 / MAX: 8.09MIN: 7.12 / MAX: 8.13

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_lcbad246810SE +/- 0.01, N = 37.747.777.787.81MIN: 7 / MAX: 8.06MIN: 7.09 / MAX: 8.11MIN: 7.1 / MAX: 8.12MIN: 7.08 / MAX: 8.17

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_ldabc246810SE +/- 0.02, N = 37.757.787.797.79MIN: 7.2 / MAX: 8.05MIN: 7.11 / MAX: 8.12MIN: 7.21 / MAX: 8.14MIN: 7.14 / MAX: 8.11

a

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.11.6
Security Notes: 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

Testing initiated at 26 March 2024 20:05 by user pts.

b

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.11.6
Security Notes: 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

Testing initiated at 26 March 2024 21:59 by user pts.

c

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.11.6
Security Notes: 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

Testing initiated at 26 March 2024 22:41 by user pts.

d

Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 7.79GHz (64 Cores / 128 Threads), Motherboard: System76 Thelio Major (FA Z5 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32GB DRAM-4800MT/s Micron MTC20F1045S1RC48BA2, Disk: 1000GB CT1000T700SSD5, Graphics: AMD Radeon Pro W7900 45GB (1760/1124MHz), Audio: AMD Device 14cc, Monitor: DELL P2415Q, Network: Aquantia AQC113C NBase-T/IEEE + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.10, Kernel: 6.5.0-26-generic (x86_64), Desktop: GNOME Shell 45.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 23.2.1-1ubuntu3.1 (LLVM 15.0.7 DRM 3.54), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105
Python Notes: Python 3.11.6
Security Notes: 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

Testing initiated at 26 March 2024 23:37 by user pts.