HDVR4-A8.9600-1

AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C+6G testing with a ASRock A320M-HDV R4.0 (P2.00 BIOS) and llvmpipe on Ubuntu 20.04 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 2312201-HERT-HDVR4A802
Jump To Table - Results

Statistics

Remove Outliers Before Calculating Averages

Graph Settings

Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C
December 12 2023
 
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):

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C

Processor: AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C+6G @ 3.10GHz (2 Cores / 4 Threads), Motherboard: ASRock A320M-HDV R4.0 (P2.00 BIOS), Chipset: AMD 15h, Memory: 3584MB, Disk: 1000GB Western Digital WDS100T2B0A, Graphics: llvmpipe, Audio: AMD Kabini HDMI/DP, Network: Realtek RTL8111/8168/8411

OS: Ubuntu 20.04, Kernel: 5.15.0-89-generic (x86_64), Desktop: GNOME Shell 3.36.9, Display Server: X Server 1.20.13, OpenGL: 4.5 Mesa 21.2.6 (LLVM 12.0.0 256 bits), Vulkan: 1.1.182, Compiler: GCC 9.4.0, File-System: ext4, Screen Resolution: 1368x768

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-9QDOt0/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x600611a
Python Notes: Python 3.8.10
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: Mitigation of untrained return thunk; SMT vulnerable + spec_rstack_overflow: 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 Retpolines IBPB: conditional STIBP: disabled RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

HDVR4-A8.9600-1OpenBenchmarking.orgPhoronix Test SuiteAMD A8-9600 RADEON R7 10 COMPUTE CORES 4C+6G @ 3.10GHz (2 Cores / 4 Threads)ASRock A320M-HDV R4.0 (P2.00 BIOS)AMD 15h3584MB1000GB Western Digital WDS100T2B0AllvmpipeAMD Kabini HDMI/DPRealtek RTL8111/8168/8411Ubuntu 20.045.15.0-89-generic (x86_64)GNOME Shell 3.36.9X Server 1.20.134.5 Mesa 21.2.6 (LLVM 12.0.0 256 bits)1.1.182GCC 9.4.0ext41368x768ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionHDVR4-A8.9600-1 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-9QDOt0/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x600611a- Python 3.8.10- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT vulnerable + spec_rstack_overflow: 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 Retpolines IBPB: conditional STIBP: disabled RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

HDVR4-A8.9600-1whisper-cpp: ggml-medium.en - 2016 State of the Uniontensorflow: CPU - 64 - VGG-16scikit-learn: Sparse Rand Projections / 100 Iterationstensorflow: CPU - 256 - GoogLeNetwhisper-cpp: ggml-small.en - 2016 State of the Uniontensorflow: CPU - 64 - ResNet-50tensorflow: CPU - 32 - VGG-16pytorch: CPU - 32 - Efficientnet_v2_ltensorflow: CPU - 512 - AlexNetpytorch: CPU - 16 - Efficientnet_v2_ltensorflow: CPU - 16 - VGG-16pytorch: CPU - 64 - Efficientnet_v2_lpytorch: CPU - 256 - Efficientnet_v2_lpytorch: CPU - 512 - Efficientnet_v2_lscikit-learn: SAGAscikit-learn: GLMwhisper-cpp: ggml-base.en - 2016 State of the Uniontensorflow: CPU - 32 - ResNet-50tensorflow: CPU - 256 - AlexNetcaffe: GoogleNet - CPU - 1000pytorch: CPU - 32 - ResNet-50pytorch: CPU - 16 - ResNet-152pytorch: CPU - 32 - ResNet-152scikit-learn: TSNE MNIST Datasetpytorch: CPU - 64 - ResNet-152pytorch: CPU - 256 - ResNet-152pytorch: CPU - 512 - ResNet-152scikit-learn: Lassoplaidml: No - Inference - VGG16 - CPUtensorflow: CPU - 64 - GoogLeNetpytorch: CPU - 256 - ResNet-50plaidml: No - Inference - ResNet 50 - CPUscikit-learn: Covertype Dataset Benchmarktensorflow: CPU - 16 - ResNet-50scikit-learn: Plot Lasso Pathscikit-learn: Hist Gradient Boosting Threadingcaffe: AlexNet - CPU - 1000pytorch: CPU - 1 - Efficientnet_v2_lscikit-learn: Plot Polynomial Kernel Approximationnumenta-nab: KNN CADscikit-learn: Plot Hierarchicalscikit-learn: Kernel PCA Solvers / Time vs. N Samplesnumenta-nab: Earthgecko Skylinescikit-learn: Plot Singular Value Decompositionmnn: inception-v3mnn: mobilenet-v1-1.0mnn: MobileNetV2_224mnn: SqueezeNetV1.0mnn: resnet-v2-50mnn: squeezenetv1.1mnn: mobilenetV3mnn: nasnetpytorch: CPU - 16 - ResNet-50scikit-learn: Plot Neighborspytorch: CPU - 64 - ResNet-50pytorch: CPU - 512 - ResNet-50deepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Streamtensorflow: CPU - 32 - GoogLeNetscikit-learn: LocalOutlierFactorncnn: Vulkan GPU - FastestDetncnn: Vulkan GPU - vision_transformerncnn: Vulkan GPU - regnety_400mncnn: Vulkan GPU - squeezenet_ssdncnn: Vulkan GPU - yolov4-tinyncnn: Vulkan GPU - resnet50ncnn: Vulkan GPU - alexnetncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - vgg16ncnn: Vulkan GPU - googlenetncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - mnasnetncnn: Vulkan GPU - shufflenet-v2ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - mobilenetncnn: CPU - FastestDetncnn: CPU - vision_transformerncnn: CPU - regnety_400mncnn: CPU - squeezenet_ssdncnn: CPU - yolov4-tinyncnn: CPU - resnet50ncnn: CPU - alexnetncnn: CPU - resnet18ncnn: CPU - vgg16ncnn: CPU - googlenetncnn: CPU - blazefacencnn: CPU - efficientnet-b0ncnn: CPU - mnasnetncnn: CPU - shufflenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - mobilenetpytorch: CPU - 1 - ResNet-152tensorflow: CPU - 64 - AlexNetscikit-learn: Feature Expansionstnn: CPU - DenseNetscikit-learn: Treescikit-learn: Hist Gradient Boostingopencv: DNN - Deep Neural Networkscikit-learn: Hist Gradient Boosting Higgs Bosonnumpy: scikit-learn: SGD Regressionscikit-learn: Plot OMP vs. LARScaffe: GoogleNet - CPU - 200scikit-learn: Sample Without Replacementscikit-learn: Kernel PCA Solvers / Time vs. N Componentstensorflow: CPU - 16 - GoogLeNettensorflow: CPU - 32 - AlexNetlczero: BLASdeepspeech: CPUnumenta-nab: Bayesian Changepointopenvino: Handwritten English Recognition FP16 - CPUopenvino: Handwritten English Recognition FP16 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUopenvino: Face Detection Retail FP16-INT8 - CPUtensorflow-lite: Mobilenet Floatscikit-learn: Sparsifypytorch: CPU - 1 - ResNet-50onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUnumenta-nab: Contextual Anomaly Detector OSEscikit-learn: Plot Wardscikit-learn: Hist Gradient Boosting Adulttensorflow: CPU - 16 - AlexNetdeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamscikit-learn: MNIST Datasetcaffe: GoogleNet - CPU - 100mlpack: scikit_icascikit-learn: Text Vectorizerscaffe: AlexNet - CPU - 200onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUmlpack: scikit_svmscikit-learn: 20 Newsgroups / Logistic Regressionscikit-learn: Plot Incremental PCAonnx: super-resolution-10 - CPU - Parallelonnx: super-resolution-10 - CPU - Parallelnumenta-nab: Relative Entropyonnx: fcn-resnet101-11 - CPU - Parallelonnx: fcn-resnet101-11 - CPU - Parallelcaffe: AlexNet - CPU - 100onnx: fcn-resnet101-11 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardonnx: bertsquad-12 - CPU - Parallelonnx: bertsquad-12 - CPU - Parallelonnx: bertsquad-12 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Parallelonnx: ArcFace ResNet-100 - CPU - Parallelonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: Faster R-CNN R-50-FPN-int8 - CPU - Standardonnx: GPT-2 - CPU - Parallelonnx: GPT-2 - CPU - Parallelonnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallelonnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallelonnx: GPT-2 - CPU - Standardonnx: GPT-2 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Parallelonnx: CaffeNet 12-int8 - CPU - Parallelonnx: ResNet50 v1-12-int8 - CPU - Parallelonnx: ResNet50 v1-12-int8 - CPU - Parallelonnx: CaffeNet 12-int8 - CPU - Standardonnx: CaffeNet 12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: ResNet50 v1-12-int8 - CPU - Standardonnx: super-resolution-10 - CPU - Standardonnx: super-resolution-10 - CPU - Standarddeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamopenvino: Handwritten English Recognition FP16-INT8 - CPUopenvino: Handwritten English Recognition FP16-INT8 - CPUscikit-learn: Hist Gradient Boosting Categorical Onlydeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamopenvino: Face Detection FP16 - CPUopenvino: Face Detection FP16 - CPUopenvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUtensorflow-lite: Inception V4tensorflow-lite: Inception ResNet V2openvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Road Segmentation ADAS FP16-INT8 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Road Segmentation ADAS FP16 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUopenvino: Vehicle Detection FP16-INT8 - CPUrbenchmark: openvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16 - CPUtensorflow-lite: NASNet Mobiledeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUtensorflow-lite: SqueezeNettensorflow-lite: Mobilenet Quantopenvino: Face Detection Retail FP16 - CPUopenvino: Face Detection Retail FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamnumenta-nab: Windowed Gaussiantnn: CPU - MobileNet v2deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamtnn: CPU - SqueezeNet v1.1deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamrnnoise: onednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUtnn: CPU - SqueezeNet v2onednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C29217.3950.458668.5253.028252.80631.020.530.5912.230.590.610.600.600.601919.8121877.5352498.666581.5012.0223185502.210.910.931425.7880.920.920.931312.0641.334.582.171.60965.0251.47835.044838.41010791131.10761.9591009.020722.267720.168953.649692.635163.10320.46117.25332.409130.58117.6256.95846.6122.24599.8812.212.231089.99070.91814.53559.53822.621140.7133.2068.80139.80135.9842.2561.47335.2375.415.3239.2525.6216.5322.4531.44100.5322.261141.9333.6868.66139.59135.4842.2960.75333.5675.155.3339.5025.6316.3222.5831.61100.211.6210.54491.6328874.345103.793414.916106960309.508122.42358.670355.351457100338.633329.1944.489.07124378.78422325.849593.053.4030.7365.0826921.0226.7923.8538767.338464.938466.3272.340202.927201.9987.022245.61140.8881177.394231503223.37171.50321658821318.621158.021148.244.60126.691110.289120.1598.32539119.85613409.40.07457731073109394.740.1064431319.060.758124814.1661.22825514.1721.94502893.6521.1190085.332711.71641109.540.90128046.244821.6196281.8393.5480838.092326.2489154.2406.4832922.431944.575284.580011.822887.783911.395274.183413.4772587.133.4161.348137.265214.55888841.250.236728.060.3316.58606.3146157.30976.35612218.22240.90121041.21910.96041044.451.911042.191.92716.362.79547885419892202.739.8684.0923.78426.614.6998.9520.200.6143168.7111.8566075.72785.39760.713967.7629.4992.7621.5534007.228341.742.7046.802.90681.704.54438.382788.65160.71681351.89130.73981354.91260.738057.324656.292325.30016.1356181.78905.5004262.33917.6144261.93797.6267554.86023.6022565.80253.5275253.35703.9464120.21288.3171250.61343.989642.692646.7652119.96628.3342557.95023.409042.681536.43553.298132.510842.233923.787360.619745.541012.0047135.31290.591070.478454.1334OpenBenchmarking.org

Whisper.cpp

Whisper.cpp is a port of OpenAI's Whisper model in C/C++. Whisper.cpp is developed by Georgi Gerganov for transcribing WAV audio files to text / speech recognition. Whisper.cpp supports ARM NEON, x86 AVX, and other advanced CPU features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.4Model: ggml-medium.en - Input: 2016 State of the Unionllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C6K12K18K24K30KSE +/- 3.98, N = 329217.401. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: VGG-16llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.10130.20260.30390.40520.5065SE +/- 0.00, N = 30.45

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Sparse Random Projections / 100 Iterationsllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2K4K6K8K10KSE +/- 67.77, N = 38668.531. (F9X) gfortran options: -O0

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 256 - Model: GoogLeNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.67951.3592.03852.7183.3975SE +/- 0.03, N = 33.02

Whisper.cpp

Whisper.cpp is a port of OpenAI's Whisper model in C/C++. Whisper.cpp is developed by Georgi Gerganov for transcribing WAV audio files to text / speech recognition. Whisper.cpp supports ARM NEON, x86 AVX, and other advanced CPU features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.4Model: ggml-small.en - Input: 2016 State of the Unionllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2K4K6K8K10KSE +/- 4.12, N = 38252.811. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.22950.4590.68850.9181.1475SE +/- 0.01, N = 31.02

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: VGG-16llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.11930.23860.35790.47720.5965SE +/- 0.00, N = 30.53

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

Benchmark: SGDOneClassSVM

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

Benchmark: Isolation Forest

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 32 - Model: Efficientnet_v2_lllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.13280.26560.39840.53120.664SE +/- 0.01, N = 50.59MIN: 0.45 / MAX: 0.66

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

Benchmark: Plot Fast KMeans

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: AlexNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.02, N = 312.23

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 16 - Model: Efficientnet_v2_lllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.13280.26560.39840.53120.664SE +/- 0.00, N = 30.59MIN: 0.45 / MAX: 0.66

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: VGG-16llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.13730.27460.41190.54920.6865SE +/- 0.00, N = 30.61

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 64 - Model: Efficientnet_v2_lllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.1350.270.4050.540.675SE +/- 0.01, N = 30.60MIN: 0.44 / MAX: 0.66

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 256 - Model: Efficientnet_v2_lllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.1350.270.4050.540.675SE +/- 0.00, N = 30.60MIN: 0.45 / MAX: 0.66

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_lllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.1350.270.4050.540.675SE +/- 0.01, N = 30.60MIN: 0.45 / MAX: 0.66

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: SAGAllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C400800120016002000SE +/- 9.65, N = 31919.811. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: GLMllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C400800120016002000SE +/- 3.03, N = 31877.541. (F9X) gfortran options: -O0

Whisper.cpp

Whisper.cpp is a port of OpenAI's Whisper model in C/C++. Whisper.cpp is developed by Georgi Gerganov for transcribing WAV audio files to text / speech recognition. Whisper.cpp supports ARM NEON, x86 AVX, and other advanced CPU features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.4Model: ggml-base.en - Input: 2016 State of the Unionllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C5001000150020002500SE +/- 3.14, N = 32498.671. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.33750.6751.01251.351.6875SE +/- 0.00, N = 31.50

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 256 - Model: AlexNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.01, N = 312.02

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 1000llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C500K1000K1500K2000K2500KSE +/- 23558.22, N = 323185501. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 32 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.49730.99461.49191.98922.4865SE +/- 0.02, N = 82.21MIN: 1.46 / MAX: 2.33

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 16 - Model: ResNet-152llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.20480.40960.61440.81921.024SE +/- 0.01, N = 30.91MIN: 0.58 / MAX: 0.96

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 32 - Model: ResNet-152llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.20930.41860.62790.83721.0465SE +/- 0.01, N = 30.93MIN: 0.58 / MAX: 0.96

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: TSNE MNIST Datasetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C30060090012001500SE +/- 3.72, N = 31425.791. (F9X) gfortran options: -O0

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 64 - Model: ResNet-152llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.2070.4140.6210.8281.035SE +/- 0.00, N = 30.92MIN: 0.59 / MAX: 0.96

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 256 - Model: ResNet-152llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.2070.4140.6210.8281.035SE +/- 0.00, N = 30.92MIN: 0.58 / MAX: 0.96

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 512 - Model: ResNet-152llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.20930.41860.62790.83721.0465SE +/- 0.00, N = 30.93MIN: 0.56 / MAX: 0.96

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Lassollvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C30060090012001500SE +/- 2.47, N = 31312.061. (F9X) gfortran options: -O0

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.29930.59860.89791.19721.4965SE +/- 0.01, N = 31.33

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: GoogLeNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.03052.0613.09154.1225.1525SE +/- 0.00, N = 34.58

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 256 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.48830.97661.46491.95322.4415SE +/- 0.02, N = 52.17MIN: 1.45 / MAX: 2.3

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.360.721.081.441.8SE +/- 0.01, N = 31.60

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Covertype Dataset Benchmarkllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.44, N = 3965.031. (F9X) gfortran options: -O0

Benchmark: Isotonic / Perturbed Logarithm

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.33080.66160.99241.32321.654SE +/- 0.01, N = 31.47

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Lasso Pathllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.49, N = 3835.041. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Hist Gradient Boosting Threadingllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.98, N = 3838.411. (F9X) gfortran options: -O0

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 1000llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C200K400K600K800K1000KSE +/- 2982.40, N = 310791131. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_lllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.24750.4950.74250.991.2375SE +/- 0.01, N = 31.10MIN: 0.79 / MAX: 1.16

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Polynomial Kernel Approximationllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C160320480640800SE +/- 0.98, N = 3761.961. (F9X) gfortran options: -O0

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: KNN CADllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 6.74, N = 31009.02

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Hierarchicalllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C160320480640800SE +/- 0.36, N = 3722.271. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Kernel PCA Solvers / Time vs. N Samplesllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C160320480640800SE +/- 1.47, N = 3720.171. (F9X) gfortran options: -O0

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko Skylinellvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.84, N = 3953.65

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Singular Value Decompositionllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C150300450600750SE +/- 2.73, N = 3692.641. (F9X) gfortran options: -O0

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: inception-v3llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 0.98, N = 3163.10MIN: 160.43 / MAX: 252.411. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: mobilenet-v1-1.0llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.09, N = 320.46MIN: 20.1 / MAX: 40.61. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: MobileNetV2_224llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C48121620SE +/- 0.06, N = 317.25MIN: 16.89 / MAX: 37.421. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: SqueezeNetV1.0llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C816243240SE +/- 0.08, N = 332.41MIN: 31.77 / MAX: 53.111. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: resnet-v2-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.22, N = 3130.58MIN: 129.32 / MAX: 174.021. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: squeezenetv1.1llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C48121620SE +/- 0.12, N = 317.63MIN: 17.14 / MAX: 37.871. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: mobilenetV3llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.031, N = 36.958MIN: 6.82 / MAX: 9.381. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: nasnetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1122334455SE +/- 0.18, N = 346.61MIN: 45.85 / MAX: 67.491. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 16 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.5041.0081.5122.0162.52SE +/- 0.01, N = 32.24MIN: 1.48 / MAX: 2.34

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Neighborsllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C130260390520650SE +/- 1.93, N = 3599.881. (F9X) gfortran options: -O0

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 64 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.49730.99461.49191.98922.4865SE +/- 0.02, N = 32.21MIN: 1.41 / MAX: 2.32

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 512 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.50181.00361.50542.00722.509SE +/- 0.02, N = 32.23MIN: 1.47 / MAX: 2.34

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 10.83, N = 91089.99

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.20660.41320.61980.82641.033SE +/- 0.0086, N = 90.9181

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: GoogLeNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.01932.03863.05794.07725.0965SE +/- 0.00, N = 34.53

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: LocalOutlierFactorllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C120240360480600SE +/- 2.37, N = 3559.541. (F9X) gfortran options: -O0

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.05, N = 322.62MIN: 22.2 / MAX: 28.411. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 0.81, N = 31140.71MIN: 1116.02 / MAX: 1259.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C816243240SE +/- 0.03, N = 333.20MIN: 32.32 / MAX: 52.571. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1530456075SE +/- 0.35, N = 368.80MIN: 67.79 / MAX: 146.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinyllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.25, N = 3139.80MIN: 137.44 / MAX: 150.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.11, N = 3135.98MIN: 134.41 / MAX: 153.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.10, N = 342.25MIN: 41.49 / MAX: 44.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1428425670SE +/- 0.12, N = 361.47MIN: 60.86 / MAX: 79.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vgg16llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 1.47, N = 3335.23MIN: 328.93 / MAX: 356.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.17, N = 375.41MIN: 73.99 / MAX: 100.711. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefacellvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.1972.3943.5914.7885.985SE +/- 0.03, N = 35.32MIN: 5.19 / MAX: 8.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C918273645SE +/- 0.08, N = 339.25MIN: 38.83 / MAX: 45.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C612182430SE +/- 0.04, N = 325.62MIN: 25.19 / MAX: 58.611. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C48121620SE +/- 0.10, N = 316.53MIN: 16.14 / MAX: 36.471. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.09, N = 322.45MIN: 21.93 / MAX: 41.711. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C714212835SE +/- 0.10, N = 331.44MIN: 31.12 / MAX: 34.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mobilenetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.16, N = 3100.53MIN: 99.32 / MAX: 121.451. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: FastestDetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.11, N = 322.26MIN: 21.72 / MAX: 40.611. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vision_transformerllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 0.72, N = 31141.93MIN: 1114.61 / MAX: 1288.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: regnety_400mllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C816243240SE +/- 0.11, N = 333.68MIN: 32.55 / MAX: 40.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: squeezenet_ssdllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1530456075SE +/- 0.46, N = 368.66MIN: 67.06 / MAX: 88.531. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: yolov4-tinyllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.30, N = 3139.59MIN: 136.68 / MAX: 156.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.32, N = 3135.48MIN: 134.06 / MAX: 143.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: alexnetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.13, N = 342.29MIN: 41.47 / MAX: 57.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet18llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1428425670SE +/- 0.25, N = 360.75MIN: 59.78 / MAX: 79.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vgg16llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 1.44, N = 3333.56MIN: 326.64 / MAX: 373.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: googlenetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.12, N = 375.15MIN: 73.47 / MAX: 97.991. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: blazefacellvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.19932.39863.59794.79725.9965SE +/- 0.02, N = 35.33MIN: 5.2 / MAX: 5.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: efficientnet-b0llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C918273645SE +/- 0.04, N = 339.50MIN: 39.16 / MAX: 43.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mnasnetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C612182430SE +/- 0.04, N = 325.63MIN: 25.22 / MAX: 28.141. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: shufflenet-v2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C48121620SE +/- 0.23, N = 316.32MIN: 15.65 / MAX: 36.761. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v3-v3 - Model: mobilenet-v3llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.11, N = 322.58MIN: 22.01 / MAX: 65.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v2-v2 - Model: mobilenet-v2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C714212835SE +/- 0.13, N = 331.61MIN: 31.19 / MAX: 35.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mobilenetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.14, N = 3100.21MIN: 99.08 / MAX: 122.541. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 1 - Model: ResNet-152llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.36450.7291.09351.4581.8225SE +/- 0.01, N = 31.62MIN: 1.04 / MAX: 1.7

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: AlexNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.01, N = 310.54

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Feature Expansionsllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C110220330440550SE +/- 0.21, N = 3491.631. (F9X) gfortran options: -O0

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: DenseNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2K4K6K8K10KSE +/- 65.94, N = 38874.35MIN: 8585.49 / MAX: 9341.041. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

Benchmark: scikit_qda

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Benchmark: scikit_linearridgeregression

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Treellvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.84, N = 15103.791. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Hist Gradient Boostingllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C90180270360450SE +/- 0.77, N = 3414.921. (F9X) gfortran options: -O0

Benchmark: Isotonic / Logistic

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: DNN - Deep Neural Networkllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20K40K60K80K100KSE +/- 2616.66, N = 151069601. (CXX) g++ options: -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -ldl -lm -lpthread -lrt

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Hist Gradient Boosting Higgs Bosonllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 2.43, N = 3309.511. (F9X) gfortran options: -O0

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy Benchmarkllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.39, N = 3122.42

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: SGD Regressionllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C80160240320400SE +/- 0.51, N = 3358.671. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot OMP vs. LARSllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C80160240320400SE +/- 4.35, N = 3355.351. (F9X) gfortran options: -O0

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 200llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C100K200K300K400K500KSE +/- 498.93, N = 34571001. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Sample Without Replacementllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 2.47, N = 3338.631. (F9X) gfortran options: -O0

Benchmark: Isotonic / Pathological

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Kernel PCA Solvers / Time vs. N Componentsllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 0.55, N = 3329.191. (F9X) gfortran options: -O0

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: GoogLeNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.0082.0163.0244.0325.04SE +/- 0.01, N = 34.48

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: AlexNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.00, N = 39.07

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 1.20, N = 31241. (CXX) g++ options: -flto -pthread

DeepSpeech

Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDeepSpeech 0.6Acceleration: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C80160240320400SE +/- 0.80, N = 3378.78

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Bayesian Changepointllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 3.10, N = 3325.85

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C130260390520650SE +/- 12.86, N = 15593.05MIN: 368.67 / MAX: 676.181. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.7651.532.2953.063.825SE +/- 0.08, N = 153.401. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C714212835SE +/- 0.25, N = 1530.73MIN: 17.23 / MAX: 53.21. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1530456075SE +/- 0.53, N = 1565.081. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet Floatllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C6K12K18K24K30KSE +/- 228.19, N = 1526921.0

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Sparsifyllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C50100150200250SE +/- 0.05, N = 3226.791. (F9X) gfortran options: -O0

PyTorch

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.1Device: CPU - Batch Size: 1 - Model: ResNet-50llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.86631.73262.59893.46524.3315SE +/- 0.02, N = 33.85MIN: 2.63 / MAX: 4.09

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C8K16K24K32K40KSE +/- 123.68, N = 338767.3MIN: 38319.31. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C8K16K24K32K40KSE +/- 111.64, N = 338464.9MIN: 38219.31. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C8K16K24K32K40KSE +/- 25.25, N = 338466.3MIN: 38310.51. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Contextual Anomaly Detector OSEllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C60120180240300SE +/- 0.36, N = 3272.34

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Wardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 0.33, N = 3202.931. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Hist Gradient Boosting Adultllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 2.73, N = 3202.001. (F9X) gfortran options: -O0

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: AlexNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.03, N = 37.02

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

Benchmark: Plot Non-Negative Matrix Factorization

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: KeyError:

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C5001000150020002500SE +/- 11.37, N = 32245.61

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.19980.39960.59940.79920.999SE +/- 0.0040, N = 30.8881

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: MNIST Datasetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 0.10, N = 3177.391. (F9X) gfortran options: -O0

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 100llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C50K100K150K200K250KSE +/- 2399.23, N = 32315031. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icallvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C50100150200250SE +/- 0.45, N = 3223.37

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Text Vectorizersllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 0.44, N = 3171.501. (F9X) gfortran options: -O0

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 200llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C50K100K150K200K250KSE +/- 882.15, N = 32165881. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C5K10K15K20K25KSE +/- 32.23, N = 321318.6MIN: 21195.41. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C5K10K15K20K25KSE +/- 31.33, N = 321158.0MIN: 21017.11. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C5K10K15K20K25KSE +/- 10.14, N = 321148.2MIN: 21028.31. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.34, N = 1044.60

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: 20 Newsgroups / Logistic Regressionllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.05, N = 3126.691. (F9X) gfortran options: -O0

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Plot Incremental PCAllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.25, N = 3110.291. (F9X) gfortran options: -O0

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: super-resolution-10 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 1.40, N = 4120.161. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: super-resolution-10 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.09724, N = 48.325391. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: super-resolution-10 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 1.40, N = 4120.161. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Relative Entropyllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 1.36, N = 3119.86

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: fcn-resnet101-11 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3K6K9K12K15KSE +/- 55.92, N = 313409.41. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: fcn-resnet101-11 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.01680.03360.05040.06720.084SE +/- 0.0003114, N = 30.07457731. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: fcn-resnet101-11 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3K6K9K12K15KSE +/- 55.92, N = 313409.41. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 100llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20K40K60K80K100KSE +/- 551.70, N = 31073101. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: fcn-resnet101-11 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2K4K6K8K10KSE +/- 10.14, N = 39394.741. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: fcn-resnet101-11 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.02390.04780.07170.09560.1195SE +/- 0.000115, N = 30.1064431. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: fcn-resnet101-11 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2K4K6K8K10KSE +/- 10.14, N = 39394.741. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

Device: CPU - Batch Size: 512 - Model: VGG-16

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: Fatal Python error: Aborted

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: bertsquad-12 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C30060090012001500SE +/- 4.38, N = 31319.061. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: bertsquad-12 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.17060.34120.51180.68240.853SE +/- 0.002525, N = 30.7581241. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: bertsquad-12 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C30060090012001500SE +/- 4.38, N = 31319.061. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: bertsquad-12 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.30, N = 3814.171. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: bertsquad-12 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.27640.55280.82921.10561.382SE +/- 0.00196, N = 31.228251. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: bertsquad-12 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.30, N = 3814.171. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C110220330440550SE +/- 3.30, N = 3514.171. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.43760.87521.31281.75042.188SE +/- 0.01254, N = 31.945021. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C110220330440550SE +/- 3.30, N = 3514.171. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 0.67, N = 3893.651. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.25180.50360.75541.00721.259SE +/- 0.00084, N = 31.119001. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 0.67, N = 3893.651. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.12, N = 385.331. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.02, N = 311.721. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.12, N = 385.331. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 2.36, N = 31109.541. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.20280.40560.60840.81121.014SE +/- 0.001919, N = 30.9012801. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 2.36, N = 31109.541. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.04, N = 346.241. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.02, N = 321.621. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: GPT-2 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.04, N = 346.241. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ArcFace ResNet-100 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C60120180240300SE +/- 0.20, N = 3281.841. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: ArcFace ResNet-100 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.79831.59662.39493.19323.9915SE +/- 0.00258, N = 33.548081. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ArcFace ResNet-100 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C60120180240300SE +/- 0.20, N = 3281.841. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C918273645SE +/- 0.05, N = 338.091. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C612182430SE +/- 0.03, N = 326.251. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C918273645SE +/- 0.05, N = 338.091. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.40, N = 3154.241. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.01700, N = 36.483291. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallelllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.40, N = 3154.241. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.05, N = 322.431. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.09, N = 344.581. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: CaffeNet 12-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.05, N = 322.431. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.05, N = 384.581. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.01, N = 311.821. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.05, N = 384.581. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: super-resolution-10 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 1.18, N = 387.781. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

OpenBenchmarking.orgInferences Per Second, More Is BetterONNX Runtime 1.14Model: super-resolution-10 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.15, N = 311.401. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread
OpenBenchmarking.orgInference Time Cost (ms), Fewer Is BetterONNX Runtime 1.14Model: super-resolution-10 - Device: CPU - Executor: Standardllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 1.18, N = 387.781. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt -lpthread -pthread

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1632486480SE +/- 0.04, N = 374.18

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.01, N = 313.48

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C130260390520650SE +/- 7.12, N = 4587.13MIN: 363.64 / MAX: 654.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Handwritten English Recognition FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.76731.53462.30193.06923.8365SE +/- 0.04, N = 43.411. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.2.2Benchmark: Hist Gradient Boosting Categorical Onlyllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1428425670SE +/- 0.19, N = 361.351. (F9X) gfortran options: -O0

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.10, N = 3137.27

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C48121620SE +/- 0.01, N = 314.56

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2K4K6K8K10KSE +/- 5.44, N = 38841.25MIN: 8808.56 / MAX: 8886.061. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.05180.10360.15540.20720.259SE +/- 0.00, N = 30.231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C14002800420056007000SE +/- 12.52, N = 36728.06MIN: 6553.59 / MAX: 7002.511. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.06750.1350.20250.270.3375SE +/- 0.00, N = 30.31. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 3.55, N = 3316.59

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0711, N = 36.3146

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.53, N = 3157.31

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0214, N = 36.3561

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C5001000150020002500SE +/- 0.75, N = 32218.22

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.20280.40560.60840.81121.014SE +/- 0.0003, N = 30.9012

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

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

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

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

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.84, N = 31041.22

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.21610.43220.64830.86441.0805SE +/- 0.0017, N = 30.9604

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Detection FP32 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.28, N = 31044.45MIN: 970.14 / MAX: 1115.691. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP32 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.42980.85961.28941.71922.149SE +/- 0.00, N = 31.911. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C2004006008001000SE +/- 1.70, N = 31042.19MIN: 990.71 / MAX: 1084.321. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.4320.8641.2961.7282.16SE +/- 0.00, N = 31.921. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C150300450600750SE +/- 1.68, N = 3716.36MIN: 663.2 / MAX: 742.361. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Machine Translation EN To DE FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.62781.25561.88342.51123.139SE +/- 0.00, N = 32.791. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception V4llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C120K240K360K480K600KSE +/- 5686.85, N = 3547885

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception ResNet V2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C90K180K270K360K450KSE +/- 439.95, N = 3419892

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 1.03, N = 3202.73MIN: 172.53 / MAX: 241.671. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.05, N = 39.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 1.13, N = 384.09MIN: 43.23 / MAX: 147.081. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C612182430SE +/- 0.32, N = 323.781. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C90180270360450SE +/- 1.31, N = 3426.61MIN: 366.94 / MAX: 466.481. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Road Segmentation ADAS FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.05532.11063.16594.22125.2765SE +/- 0.01, N = 34.691. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.80, N = 398.95MIN: 63.4 / MAX: 121.431. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.16, N = 320.201. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

R Benchmark

This test is a quick-running survey of general R performance Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterR Benchmarkllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.13820.27640.41460.55280.691SE +/- 0.0070, N = 30.61431. R scripting front-end version 3.6.3 (2020-02-29)

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 0.99, N = 3168.71MIN: 119.34 / MAX: 195.231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Vehicle Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.07, N = 311.851. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: NASNet Mobilellvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C14K28K42K56K70KSE +/- 427.94, N = 366075.7

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C6001200180024003000SE +/- 3.11, N = 32785.40

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.16060.32120.48180.64240.803SE +/- 0.0019, N = 30.7139

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1530456075SE +/- 0.29, N = 367.76MIN: 39.77 / MAX: 88.071. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C714212835SE +/- 0.13, N = 329.491. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.25, N = 392.76MIN: 67.15 / MAX: 134.831. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Weld Porosity Detection FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C510152025SE +/- 0.06, N = 321.551. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: SqueezeNetllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C7K14K21K28K35KSE +/- 45.90, N = 334007.2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet Quantllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C6K12K18K24K30KSE +/- 20.03, N = 328341.7

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.09, N = 342.70MIN: 20.92 / MAX: 63.971. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Face Detection Retail FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1122334455SE +/- 0.10, N = 346.801. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.65251.3051.95752.613.2625SE +/- 0.01, N = 32.90MIN: 1.7 / MAX: 21.021. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C150300450600750SE +/- 2.10, N = 3681.701. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.02152.0433.06454.0865.1075SE +/- 0.01, N = 34.54MIN: 2.6 / MAX: 35.331. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2023.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C90180270360450SE +/- 1.43, N = 3438.381. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared -ldl

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C6001200180024003000SE +/- 1.19, N = 32788.65

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.16130.32260.48390.64520.8065SE +/- 0.0003, N = 30.7168

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C30060090012001500SE +/- 7.57, N = 31351.89

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.16650.3330.49950.6660.8325SE +/- 0.0042, N = 30.7398

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C30060090012001500SE +/- 1.17, N = 31354.91

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.16610.33220.49830.66440.8305SE +/- 0.0006, N = 30.7380

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Windowed Gaussianllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1326395265SE +/- 0.27, N = 357.32

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

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

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status.

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: MobileNet v2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C140280420560700SE +/- 1.60, N = 3656.29MIN: 647.62 / MAX: 668.31. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C70140210280350SE +/- 0.05, N = 3325.30

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0051, N = 36.1356

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C4080120160200SE +/- 0.23, N = 3181.79

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1.23762.47523.71284.95046.188SE +/- 0.0069, N = 35.5004

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

Device: CPU - Batch Size: 256 - Model: VGG-16

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: Fatal Python error: Aborted

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C60120180240300SE +/- 0.82, N = 3262.34

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0256, N = 37.6144

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C60120180240300SE +/- 0.37, N = 3261.94

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0155, N = 37.6267

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C120240360480600SE +/- 1.70, N = 3554.86

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.81051.6212.43153.2424.0525SE +/- 0.0115, N = 33.6022

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C120240360480600SE +/- 1.23, N = 3565.80

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.79371.58742.38113.17483.9685SE +/- 0.0022, N = 33.5275

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C60120180240300SE +/- 0.11, N = 3253.36

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.88791.77582.66373.55164.4395SE +/- 0.0017, N = 33.9464

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.12, N = 3120.21

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0084, N = 38.3171

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C50100150200250SE +/- 0.34, N = 3250.61

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C0.89771.79542.69313.59084.4885SE +/- 0.0054, N = 33.9896

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.19, N = 342.69

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1122334455SE +/- 0.21, N = 346.77

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.12, N = 3119.97

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Baseline - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C246810SE +/- 0.0086, N = 38.3342

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v1.1llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C120240360480600SE +/- 2.06, N = 3557.95MIN: 553.57 / MAX: 564.291. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C612182430SE +/- 0.03, N = 323.41

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.6Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Streamllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.05, N = 342.68

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C816243240SE +/- 0.07, N = 336.441. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

Benchmark: RCV1 Logreg Convergencet

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: IndexError: list index out of range

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1224364860SE +/- 0.30, N = 353.30MIN: 50.671. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C816243240SE +/- 0.05, N = 332.51MIN: 31.991. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.27, N = 342.23MIN: 40.721. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C612182430SE +/- 0.13, N = 323.79MIN: 23.161. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1428425670SE +/- 0.80, N = 1260.62MIN: 55.571. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1020304050SE +/- 0.10, N = 345.54MIN: 44.651. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C3691215SE +/- 0.04, N = 312.00MIN: 11.721. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.3Target: CPU - Model: SqueezeNet v2llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C306090120150SE +/- 0.52, N = 3135.31MIN: 132.96 / MAX: 140.411. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -fvisibility=default -O3 -rdynamic -ldl

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C20406080100SE +/- 0.17, N = 390.59MIN: 89.231. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1632486480SE +/- 0.22, N = 370.48MIN: 69.031. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.3Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUllvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C1224364860SE +/- 0.06, N = 354.13MIN: 51.761. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

Benchmark: Plot Parallel Pairwise

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: numpy.core._exceptions.MemoryError: Unable to allocate 74.5 GiB for an array with shape (100000, 100000) and data type float64

AI Benchmark Alpha

AI Benchmark Alpha is a Python library for evaluating artificial intelligence (AI) performance on diverse hardware platforms and relies upon the TensorFlow machine learning library. Learn more via the OpenBenchmarking.org test page.

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. E: AttributeError: module 'numpy' has no attribute 'typeDict'

Scikit-Learn

Scikit-learn is a Python module for machine learning built on NumPy, SciPy, and is BSD-licensed. Learn more via the OpenBenchmarking.org test page.

Benchmark: Glmnet

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: ModuleNotFoundError: No module named 'glmnet.elastic_net'

spaCy

The spaCy library is an open-source solution for advanced neural language processing (NLP). The spaCy library leverages Python and is a leading neural language processing solution. This test profile times the spaCy CPU performance with various models. Learn more via the OpenBenchmarking.org test page.

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: TypeError: issubclass() arg 1 must be a class

ECP-CANDLE

The CANDLE benchmark codes implement deep learning architectures relevant to problems in cancer. These architectures address problems at different biological scales, specifically problems at the molecular, cellular and population scales. Learn more via the OpenBenchmarking.org test page.

Benchmark: P1B2

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. E: ImportError: initialization failed

Benchmark: P3B1

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. E: ImportError: initialization failed

Benchmark: P3B2

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. E: ImportError: initialization failed

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Model Zoo. Learn more via the OpenBenchmarking.org test page.

Model: yolov4 - Device: CPU - Executor: Standard

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: onnxruntime/onnxruntime/test/onnx/onnx_model_info.cc:45 void OnnxModelInfo::InitOnnxModelInfo(const PATH_CHAR_TYPE*) open file "yolov4/yolov4.onnx" failed: No such file or directory

Model: yolov4 - Device: CPU - Executor: Parallel

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test quit with a non-zero exit status. The test quit with a non-zero exit status. The test quit with a non-zero exit status. E: onnxruntime/onnxruntime/test/onnx/onnx_model_info.cc:45 void OnnxModelInfo::InitOnnxModelInfo(const PATH_CHAR_TYPE*) open file "yolov4/yolov4.onnx" failed: No such file or directory

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI toolkit. Learn more via the OpenBenchmarking.org test page.

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C: The test run did not produce a result. The test run did not produce a result. The test run did not produce a result.

272 Results Shown

Whisper.cpp
TensorFlow
Scikit-Learn
TensorFlow
Whisper.cpp
TensorFlow:
  CPU - 64 - ResNet-50
  CPU - 32 - VGG-16
PyTorch
TensorFlow
PyTorch
TensorFlow
PyTorch:
  CPU - 64 - Efficientnet_v2_l
  CPU - 256 - Efficientnet_v2_l
  CPU - 512 - Efficientnet_v2_l
Scikit-Learn:
  SAGA
  GLM
Whisper.cpp
TensorFlow:
  CPU - 32 - ResNet-50
  CPU - 256 - AlexNet
Caffe
PyTorch:
  CPU - 32 - ResNet-50
  CPU - 16 - ResNet-152
  CPU - 32 - ResNet-152
Scikit-Learn
PyTorch:
  CPU - 64 - ResNet-152
  CPU - 256 - ResNet-152
  CPU - 512 - ResNet-152
Scikit-Learn
PlaidML
TensorFlow
PyTorch
PlaidML
Scikit-Learn
TensorFlow
Scikit-Learn:
  Plot Lasso Path
  Hist Gradient Boosting Threading
Caffe
PyTorch
Scikit-Learn
Numenta Anomaly Benchmark
Scikit-Learn:
  Plot Hierarchical
  Kernel PCA Solvers / Time vs. N Samples
Numenta Anomaly Benchmark
Scikit-Learn
Mobile Neural Network:
  inception-v3
  mobilenet-v1-1.0
  MobileNetV2_224
  SqueezeNetV1.0
  resnet-v2-50
  squeezenetv1.1
  mobilenetV3
  nasnet
PyTorch
Scikit-Learn
PyTorch:
  CPU - 64 - ResNet-50
  CPU - 512 - ResNet-50
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering - Synchronous Single-Stream:
    ms/batch
    items/sec
TensorFlow
Scikit-Learn
NCNN:
  Vulkan GPU - FastestDet
  Vulkan GPU - vision_transformer
  Vulkan GPU - regnety_400m
  Vulkan GPU - squeezenet_ssd
  Vulkan GPU - yolov4-tiny
  Vulkan GPU - resnet50
  Vulkan GPU - alexnet
  Vulkan GPU - resnet18
  Vulkan GPU - vgg16
  Vulkan GPU - googlenet
  Vulkan GPU - blazeface
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - mnasnet
  Vulkan GPU - shufflenet-v2
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU-v2-v2 - mobilenet-v2
  Vulkan GPU - mobilenet
  CPU - FastestDet
  CPU - vision_transformer
  CPU - regnety_400m
  CPU - squeezenet_ssd
  CPU - yolov4-tiny
  CPU - resnet50
  CPU - alexnet
  CPU - resnet18
  CPU - vgg16
  CPU - googlenet
  CPU - blazeface
  CPU - efficientnet-b0
  CPU - mnasnet
  CPU - shufflenet-v2
  CPU-v3-v3 - mobilenet-v3
  CPU-v2-v2 - mobilenet-v2
  CPU - mobilenet
PyTorch
TensorFlow
Scikit-Learn
TNN
Scikit-Learn:
  Tree
  Hist Gradient Boosting
OpenCV
Scikit-Learn
Numpy Benchmark
Scikit-Learn:
  SGD Regression
  Plot OMP vs. LARS
Caffe
Scikit-Learn:
  Sample Without Replacement
  Kernel PCA Solvers / Time vs. N Components
TensorFlow:
  CPU - 16 - GoogLeNet
  CPU - 32 - AlexNet
LeelaChessZero
DeepSpeech
Numenta Anomaly Benchmark
OpenVINO:
  Handwritten English Recognition FP16 - CPU:
    ms
    FPS
  Face Detection Retail FP16-INT8 - CPU:
    ms
    FPS
TensorFlow Lite
Scikit-Learn
PyTorch
oneDNN:
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
  Recurrent Neural Network Training - u8s8f32 - CPU
  Recurrent Neural Network Training - f32 - CPU
Numenta Anomaly Benchmark
Scikit-Learn:
  Plot Ward
  Hist Gradient Boosting Adult
TensorFlow
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Scikit-Learn
Caffe
Mlpack Benchmark
Scikit-Learn
Caffe
oneDNN:
  Recurrent Neural Network Inference - bf16bf16bf16 - CPU
  Recurrent Neural Network Inference - f32 - CPU
  Recurrent Neural Network Inference - u8s8f32 - CPU
Mlpack Benchmark
Scikit-Learn:
  20 Newsgroups / Logistic Regression
  Plot Incremental PCA
ONNX Runtime:
  super-resolution-10 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
Numenta Anomaly Benchmark
ONNX Runtime:
  fcn-resnet101-11 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
Caffe
ONNX Runtime:
  fcn-resnet101-11 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  bertsquad-12 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  bertsquad-12 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  ArcFace ResNet-100 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  Faster R-CNN R-50-FPN-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  GPT-2 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  Faster R-CNN R-50-FPN-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  GPT-2 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  ArcFace ResNet-100 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  CaffeNet 12-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  ResNet50 v1-12-int8 - CPU - Parallel:
    Inference Time Cost (ms)
    Inferences Per Second
  CaffeNet 12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  ResNet50 v1-12-int8 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
  super-resolution-10 - CPU - Standard:
    Inference Time Cost (ms)
    Inferences Per Second
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenVINO:
  Handwritten English Recognition FP16-INT8 - CPU:
    ms
    FPS
Scikit-Learn
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
OpenVINO:
  Face Detection FP16 - CPU:
    ms
    FPS
  Face Detection FP16-INT8 - CPU:
    ms
    FPS
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
  CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenVINO:
  Person Detection FP32 - CPU:
    ms
    FPS
  Person Detection FP16 - CPU:
    ms
    FPS
  Machine Translation EN To DE FP16 - CPU:
    ms
    FPS
TensorFlow Lite:
  Inception V4
  Inception ResNet V2
OpenVINO:
  Road Segmentation ADAS FP16-INT8 - CPU:
    ms
    FPS
  Person Vehicle Bike Detection FP16 - CPU:
    ms
    FPS
  Road Segmentation ADAS FP16 - CPU:
    ms
    FPS
  Vehicle Detection FP16-INT8 - CPU:
    ms
    FPS
R Benchmark
OpenVINO:
  Vehicle Detection FP16 - CPU:
    ms
    FPS
TensorFlow Lite
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream:
    ms/batch
    items/sec
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU:
    ms
    FPS
  Weld Porosity Detection FP16 - CPU:
    ms
    FPS
TensorFlow Lite:
  SqueezeNet
  Mobilenet Quant
OpenVINO:
  Face Detection Retail FP16 - CPU:
    ms
    FPS
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU:
    ms
    FPS
  Age Gender Recognition Retail 0013 FP16 - CPU:
    ms
    FPS
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    ms/batch
    items/sec
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
Numenta Anomaly Benchmark
TNN
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    ms/batch
    items/sec
  ResNet-50, Baseline - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  CV Detection, YOLOv5s COCO - Synchronous Single-Stream:
    ms/batch
    items/sec
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    ms/batch
    items/sec
  CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
  ResNet-50, Sparse INT8 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  ResNet-50, Baseline - Synchronous Single-Stream:
    ms/batch
    items/sec
TNN
Neural Magic DeepSparse:
  ResNet-50, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
RNNoise
oneDNN:
  Deconvolution Batch shapes_1d - f32 - CPU
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  IP Shapes 1D - f32 - CPU
  IP Shapes 1D - u8s8f32 - CPU
  Deconvolution Batch shapes_3d - u8s8f32 - CPU
  IP Shapes 3D - f32 - CPU
  IP Shapes 3D - u8s8f32 - CPU
TNN
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
  Deconvolution Batch shapes_3d - f32 - CPU

llvmpipe - AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C

Processor: AMD A8-9600 RADEON R7 10 COMPUTE CORES 4C+6G @ 3.10GHz (2 Cores / 4 Threads), Motherboard: ASRock A320M-HDV R4.0 (P2.00 BIOS), Chipset: AMD 15h, Memory: 3584MB, Disk: 1000GB Western Digital WDS100T2B0A, Graphics: llvmpipe, Audio: AMD Kabini HDMI/DP, Network: Realtek RTL8111/8168/8411

OS: Ubuntu 20.04, Kernel: 5.15.0-89-generic (x86_64), Desktop: GNOME Shell 3.36.9, Display Server: X Server 1.20.13, OpenGL: 4.5 Mesa 21.2.6 (LLVM 12.0.0 256 bits), Vulkan: 1.1.182, Compiler: GCC 9.4.0, File-System: ext4, Screen Resolution: 1368x768

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-9QDOt0/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Processor Notes: Scaling Governor: acpi-cpufreq ondemand (Boost: Enabled) - CPU Microcode: 0x600611a
Python Notes: Python 3.8.10
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: Mitigation of untrained return thunk; SMT vulnerable + spec_rstack_overflow: 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 Retpolines IBPB: conditional STIBP: disabled RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 12 December 2023 15:12 by user hertz.