TensorFlow Lite AMD Ryzen Threadripper

AMD Ryzen Threadripper 3960X 24-Core testing with a MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS) and Sapphire AMD Radeon RX 5500/5500M / Pro 5500M 4GB on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2008237-FI-TENSORFLO23.

TensorFlow Lite AMD Ryzen ThreadripperProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionTR 3960XVet 2Vet 3AMD Ryzen Threadripper 3960X 24-Core @ 3.80GHz (24 Cores / 48 Threads)MSI Creator TRX40 (MS-7C59) v1.0 (1.12N1 BIOS)AMD Starship/Matisse32GB1000GB Sabrent Rocket 4.0 1TBSapphire AMD Radeon RX 5500/5500M / Pro 5500M 4GB (1900/875MHz)AMD Navi 10 HDMI AudioASUS MG28UAquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.045.4.0-42-generic (x86_64)GNOME Shell 3.36.3X Server 1.20.8modesetting 1.20.84.6 Mesa 20.0.8 (LLVM 10.0.0)GCC 9.3.0ext43840x2160GNOME Shell 3.36.4OpenBenchmarking.orgProcessor Details- Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

TensorFlow Lite AMD Ryzen Threadrippertensorflow-lite: SqueezeNettensorflow-lite: Inception V4tensorflow-lite: NASNet Mobiletensorflow-lite: Mobilenet Floattensorflow-lite: Mobilenet Quanttensorflow-lite: Inception ResNet V2TR 3960XVet 2Vet 376058.7113792386858.351035.552131.1100742375628.9113615786852.551344.752306.2100739375472.6113494386655.851110.652160.51006490OpenBenchmarking.org

TensorFlow Lite

Model: SqueezeNet

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetTR 3960XVet 2Vet 316K32K48K64K80KSE +/- 97.87, N = 3SE +/- 329.17, N = 3SE +/- 295.88, N = 376058.775628.975472.6

TensorFlow Lite

Model: Inception V4

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4TR 3960XVet 2Vet 3200K400K600K800K1000KSE +/- 688.10, N = 3SE +/- 1424.48, N = 3SE +/- 881.67, N = 3113792311361571134943

TensorFlow Lite

Model: NASNet Mobile

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileTR 3960XVet 2Vet 320K40K60K80K100KSE +/- 144.60, N = 3SE +/- 118.06, N = 3SE +/- 271.25, N = 386858.386852.586655.8

TensorFlow Lite

Model: Mobilenet Float

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatTR 3960XVet 2Vet 311K22K33K44K55KSE +/- 148.03, N = 3SE +/- 73.95, N = 3SE +/- 192.51, N = 351035.551344.751110.6

TensorFlow Lite

Model: Mobilenet Quant

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantTR 3960XVet 2Vet 311K22K33K44K55KSE +/- 72.23, N = 3SE +/- 124.19, N = 3SE +/- 104.73, N = 352131.152306.252160.5

TensorFlow Lite

Model: Inception ResNet V2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2TR 3960XVet 2Vet 3200K400K600K800K1000KSE +/- 450.42, N = 3SE +/- 761.50, N = 3SE +/- 1302.93, N = 3100742310073931006490

Phoronix Test Suite v10.8.4