THIS IS A TEST INSTANCE. ALL YOUR CHANGES WILL BE LOST!!!!
...
Performance Boost with Intel MKL-DNN backend from various models on AWS EC2 instances
Category | Model | Latency batchsize=1 (ms, small is better) | Throughput batchsize=128 (fps, big is better) | ||||
w/o MKL-DNN | w/ MKL-DNN | speedup | w/o MKL-DNN | w/ MKL-DNN | speedup | ||
CNN/classification | ResNet-50 v1 | ||||||
ResNet-50 v2 | |||||||
Inception v3 | |||||||
Inception v4 | |||||||
DenseNet | |||||||
MobileNet | |||||||
VGG16 | |||||||
AlexNet | |||||||
inception-resnet v2 | |||||||
CNN/object detection | Faster R-CNN | ||||||
SSD-VGG16 | |||||||
SSD-MobileNet | |||||||
RNN | GNMT | ||||||
GAN | DCGAN |
...
Inference Accuracy
topology | top1top1 | top5 | ||
CPU w/o Intel MKL-DNN | CPU w/ Intel MKL-DNN | CPU w/o Intel MKL-DNN | CPU w/ Intel MKL-DNN | |
alexnet | ||||
resnet50_v1 | ||||
resnet50_v2 | ||||
vgg16 | ||||
densenet121 | ||||
squeezenet1.1 | ||||
mobilenet1.0 | ||||
inceptionv3 | ||||
SSD-VGG16 |