pytorch 2.2.1 ryzen

AMD Ryzen 9 7950X 16-Core testing with a ASUS ROG STRIX X670E-E GAMING WIFI (1905 BIOS) and NVIDIA GeForce RTX 3080 10GB 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 2403270-PTS-PYTORCH233
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
  28 Minutes
b
March 27
  28 Minutes
c
March 27
  28 Minutes
d
March 27
  1 Hour, 23 Minutes
Invert Hiding All Results Option
  42 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: 0xa601206
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 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (1905 BIOS), Chipset: AMD Device 14d8, Memory: 2 x 16GB DRAM-6000MT/s G Skill F5-6000J3038F16G, Disk: 2000GB Samsung SSD 980 PRO 2TB + 123GB SanDisk 3.2Gen1, Graphics: NVIDIA GeForce RTX 3080 10GB, Audio: NVIDIA GA102 HD Audio, Monitor: DELL U2723QE, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.10, Kernel: 6.7.0-060700-generic (x86_64), Desktop: GNOME Shell 45.2, Display Server: X Server 1.21.1.7, Display Driver: NVIDIA 550.54.14, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.4.89, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160

pytorch 2.2.1 ryzenOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads)ASUS ROG STRIX X670E-E GAMING WIFI (1905 BIOS)AMD Device 14d82 x 16GB DRAM-6000MT/s G Skill F5-6000J3038F16G2000GB Samsung SSD 980 PRO 2TB + 123GB SanDisk 3.2Gen1NVIDIA GeForce RTX 3080 10GBNVIDIA GA102 HD AudioDELL U2723QEIntel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 23.106.7.0-060700-generic (x86_64)GNOME Shell 45.2X Server 1.21.1.7NVIDIA 550.54.144.6.0OpenCL 3.0 CUDA 12.4.89GCC 13.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLOpenCLCompilerFile-SystemScreen ResolutionPytorch 2.2.1 Ryzen BenchmarksSystem Logs- Transparent Huge Pages: madvise- Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601206 - 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%102%104%105%PyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchPyTorchCPU - 1 - Efficientnet_v2_lCPU - 1 - ResNet-152CPU - 16 - ResNet-50CPU - 256 - ResNet-152CPU - 64 - ResNet-152CPU - 32 - Efficientnet_v2_lCPU - 512 - ResNet-152CPU - 256 - ResNet-50CPU - 64 - Efficientnet_v2_lCPU - 512 - ResNet-50CPU - 32 - ResNet-152CPU - 1 - ResNet-50CPU - 16 - Efficientnet_v2_lCPU - 512 - Efficientnet_v2_lCPU - 32 - ResNet-50CPU - 64 - ResNet-50CPU - 16 - ResNet-152CPU - 256 - Efficientnet_v2_l

pytorch 2.2.1 ryzenpytorch: 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_labcd72.7530.3349.1848.6448.1920.1048.8919.8148.4420.3020.5419.4016.5511.7911.4211.8011.7011.7672.6629.8148.6749.0247.9720.0147.4920.2648.8319.6419.9119.6916.0911.7711.7311.8711.7811.6772.2529.3249.7348.8548.5220.1648.6120.0649.3220.0619.7919.9815.7711.6211.7811.5911.7511.5671.2328.9647.6148.3448.5120.1848.3820.2448.1920.1419.7519.9316.2211.8411.7311.8811.7411.76OpenBenchmarking.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-50abcd1632486480SE +/- 0.34, N = 372.7572.6672.2571.23MIN: 69.82 / MAX: 74.29MIN: 67.96 / MAX: 74.36MIN: 57.14 / MAX: 74.16MIN: 64.96 / MAX: 73.24

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: ResNet-152abcd714212835SE +/- 0.11, N = 330.3329.8129.3228.96MIN: 23.35 / MAX: 30.76MIN: 28.63 / MAX: 30.21MIN: 28.93 / MAX: 29.59MIN: 22.78 / MAX: 29.7

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-50cabd1122334455SE +/- 0.11, N = 349.7349.1848.6747.61MIN: 48.72 / MAX: 50.57MIN: 48.1 / MAX: 49.79MIN: 46.88 / MAX: 49.75MIN: 45.13 / MAX: 48.9

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-50bcad1122334455SE +/- 0.13, N = 349.0248.8548.6448.34MIN: 47.89 / MAX: 49.85MIN: 46.2 / MAX: 49.69MIN: 47.2 / MAX: 49.77MIN: 45.85 / MAX: 49.33

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-50cdab1122334455SE +/- 0.35, N = 348.5248.5148.1947.97MIN: 47.06 / MAX: 49.51MIN: 46.37 / MAX: 49.94MIN: 45.76 / MAX: 49.14MIN: 45.74 / MAX: 49.08

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: ResNet-152dcab510152025SE +/- 0.04, N = 320.1820.1620.1020.01MIN: 19.67 / MAX: 20.58MIN: 19.7 / MAX: 20.4MIN: 19.61 / MAX: 20.33MIN: 17.16 / MAX: 20.31

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-50acdb1122334455SE +/- 0.66, N = 348.8948.6148.3847.49MIN: 46.38 / MAX: 49.85MIN: 46.63 / MAX: 49.33MIN: 45.71 / MAX: 50.11MIN: 46.32 / MAX: 48.77

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: ResNet-152bdca510152025SE +/- 0.03, N = 320.2620.2420.0619.81MIN: 19.92 / MAX: 20.55MIN: 19.62 / MAX: 20.59MIN: 19.66 / MAX: 20.42MIN: 19.34 / MAX: 20.05

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-50cbad1122334455SE +/- 0.19, N = 349.3248.8348.4448.19MIN: 47.73 / MAX: 50.31MIN: 47.52 / MAX: 50.16MIN: 46.94 / MAX: 49.13MIN: 37.3 / MAX: 49.6

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: ResNet-152adcb510152025SE +/- 0.04, N = 320.3020.1420.0619.64MIN: 20 / MAX: 20.47MIN: 19.6 / MAX: 20.53MIN: 19.3 / MAX: 20.27MIN: 19.34 / MAX: 19.92

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: ResNet-152abcd510152025SE +/- 0.13, N = 320.5419.9119.7919.75MIN: 20.16 / MAX: 20.7MIN: 19.61 / MAX: 20.22MIN: 19.52 / MAX: 19.99MIN: 19.22 / MAX: 20.44

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-152cdba510152025SE +/- 0.08, N = 319.9819.9319.6919.40MIN: 19.59 / MAX: 20.45MIN: 17.21 / MAX: 20.37MIN: 19.19 / MAX: 20.13MIN: 18.86 / MAX: 19.68

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_ladbc48121620SE +/- 0.07, N = 316.5516.2216.0915.77MIN: 14.5 / MAX: 16.83MIN: 15.75 / MAX: 16.51MIN: 15.88 / MAX: 16.28MIN: 13.93 / MAX: 16.03

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_ldabc3691215SE +/- 0.03, N = 311.8411.7911.7711.62MIN: 9.62 / MAX: 12.79MIN: 9.57 / MAX: 12.78MIN: 9.55 / MAX: 12.38MIN: 9.52 / MAX: 12.22

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_lcdba3691215SE +/- 0.03, N = 311.7811.7311.7311.42MIN: 9.57 / MAX: 12.66MIN: 9.64 / MAX: 12.74MIN: 9.61 / MAX: 12.62MIN: 9.66 / MAX: 12.53

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_ldbac3691215SE +/- 0.14, N = 311.8811.8711.8011.59MIN: 9.72 / MAX: 12.8MIN: 9.59 / MAX: 12.72MIN: 9.74 / MAX: 12.67MIN: 9.5 / MAX: 12.72

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_lbcda3691215SE +/- 0.10, N = 311.7811.7511.7411.70MIN: 9.74 / MAX: 12.5MIN: 9.7 / MAX: 12.73MIN: 9.67 / MAX: 12.77MIN: 9.54 / MAX: 12.17

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_ldabc3691215SE +/- 0.01, N = 311.7611.7611.6711.56MIN: 9.68 / MAX: 12.8MIN: 9.55 / MAX: 12.64MIN: 9.73 / MAX: 12.76MIN: 9.64 / MAX: 12.51

a

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601206
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.

b

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601206
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 27 March 2024 00:16 by user pts.

c

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601206
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 27 March 2024 00:50 by user pts.

d

Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (1905 BIOS), Chipset: AMD Device 14d8, Memory: 2 x 16GB DRAM-6000MT/s G Skill F5-6000J3038F16G, Disk: 2000GB Samsung SSD 980 PRO 2TB + 123GB SanDisk 3.2Gen1, Graphics: NVIDIA GeForce RTX 3080 10GB, Audio: NVIDIA GA102 HD Audio, Monitor: DELL U2723QE, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411

OS: Ubuntu 23.10, Kernel: 6.7.0-060700-generic (x86_64), Desktop: GNOME Shell 45.2, Display Server: X Server 1.21.1.7, Display Driver: NVIDIA 550.54.14, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.4.89, Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Processor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa601206
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 27 March 2024 01:29 by user pts.