| ResNet-50 | Full ImageNet Val | 11.54 images/s | 32.2 images/s | 2.79 |
| MobileNet-V1 | Full ImageNet Val | 49.21 images/s | 108.37 images/s | 2.2 |
Please note that [full ImageNet validation dataset](http://www.image-net.org/challenges/LSVRC/2012/nnoupb/ILSVRC2012_img_val.tar"full ImageNet validation dataset") can be downloaded by script `test_calibration.py` with `DATASET=full`.
Notes:
* The accuracy measurement requires the model with `label`.
* The INT8 theoretical speedup is 4X on Intel® Xeon® Cascadelake Server (please refer to `providing a theoretical peak compute gain of 4x int8 OPS over fp32 OPS` in [Reference](https://software.intel.com/en-us/articles/lower-numerical-precision-deep-learning-inference-and-training"Reference")).
* The INT8 theoretical speedup is 4X on Intel® Xeon® Cascadelake Server (please refer to `providing a theoretical peak compute gain of 4x int8 OPS over fp32 OPS` in [Reference](https://software.intel.com/en-us/articles/lower-numerical-precision-deep-learning-inference-and-training"Reference")). However, the actual test results at the model level will be less than 4X, and in general the average is about 2X. In addition, the calculation library optimization of batch size 1 is not as good as the large batch size.
