The following questions and answers are related to performance benchmarks published on the Performance Information documentation site.
New performance benchmarks are typically published on every major.minor
release of the Intel® Distribution of OpenVINO™ toolkit.
All of the models used are included in the toolkit's Open Model Zoo GitHub repository.
The models used in the performance benchmarks were chosen based on general adoption and usage in deployment scenarios. We're continuing to add new models that support a diverse set of workloads and usage.
CF means Caffe*, while TF means TensorFlow*.
All of the performance benchmarks were generated using the open-sourced tool within the Intel® Distribution of OpenVINO™ toolkit called benchmark_app
, which is available in both C++ and Python.
The image size used in the inference depends on the network being benchmarked. The following table shows the list of input sizes for each network model.
Model | Public Network | Task | Input Size (Height x Width) |
---|---|---|---|
faster_rcnn_resnet50_coco-TF | Faster RCNN Tf | object detection | 600x1024 |
googlenet-v1-CF | GoogLeNet_ILSVRC-2012_Caffe | classification | 224x224 |
googlenet-v3-TF | Inception v3 Tf | classification | 299x299 |
mobilenet-ssd-CF | SSD (MobileNet)_COCO-2017_Caffe | object detection | 300x300 |
mobilenet-v2-1.0-224-TF | MobileNet v2 Tf | classification | 224x224 |
mobilenet-v2-CF | Mobilenet V2 Caffe | classification | 224x224 |
resnet-101-CF | ResNet-101_ILSVRC-2012_Caffe | classification | 224x224 |
resnet-50-CF | ResNet-50_v1_ILSVRC-2012_Caffe | classification | 224x224 |
se-resnext-50-CF | Se-ResNext-50_ILSVRC-2012_Caffe | classification | 224x224 |
squeezenet1.1-CF | SqueezeNet_v1.1_ILSVRC-2012_Caffe | classification | 227x227 |
ssd300-CF | SSD (VGG-16)_VOC-2007_Caffe | object detection | 300x300 |
Intel partners with various vendors all over the world. Visit the Intel® AI: In Production Partners & Solutions Catalog for a list of Equipment Makers and the Supported Devices documentation. You can also remotely test and run models before purchasing any hardware by using Intel® DevCloud for the Edge.
We published a set of guidelines and recommendations to optimize your models available in an introductory guide and an advanced guide. For further support, please join the conversation in the Community Forum.
The benefit of low-precision optimization using the OpenVINO™ toolkit model optimizer extends beyond processors supporting VNNI through Intel® DL Boost. The reduced bit width of INT8 compared to FP32 allows Intel® CPU to process the data faster and thus offers better throughput on any converted model agnostic of the intrinsically supported low-precision optimizations within Intel® hardware. Please refer to INT8 vs. FP32 Comparison on Select Networks and Platforms for comparison on boost factors for different network models and a selection of Intel® CPU architectures, including AVX-2 with Intel® Core™ i7-8700T, and AVX-512 (VNNI) with Intel® Xeon® 5218T and Intel® Xeon® 8270.
We replaced googlenet-v1 to resnet-18-pytorch due to changes in developer usage. The public model resnet-18 is used by many developers as an Image Classification model. This pre-optimized model was also trained on the ImageNet database, similar to googlenet-v1. Both googlenet-v1 and resnet-18 will remain part of the Open Model Zoo. Developers are encouraged to utilize resnet-18-pytorch for Image Classification use cases.
For more complete information about performance and benchmark results, visit: www.intel.com/benchmarks and Optimization Notice. Legal Information.