Post Training Optimization - A Real Example of Integer Calibration | OpenVINO™ toolkit | Ep. 69
A real end-to-end example how to calibrate a model to INT8.
Operating the OpenVINO POT - Post Training Optimization Tool,
How to calibrate to INT8 using different strategies, how to measure the performance and the accuracy.
OpenVINO toolkit Github: https://github.com/openvinotoolkit/openvino/blob/master/docs/how_tos/POT_how_to_example.md
Please follow the git-hub copy & paste how-to:
https://github.com/openvinotoolkit/openvino/blob/master/docs/how_tos/POT_how_to_example.md
This is a full example
1. Installation required
2. Download a mobilenet model
3. Convert to IR format
4. Measure baseline performance and accuracy of the FP32 model
5. Calibrate to INT8 using default strategy .. and measure accuracy and performance
6. Accuracy aware quantization and compare accuracy and performance
About Intel Software:
The Intel® Developer Zone encourages and supports software developers that are developing applications for Intel hardware and software products. The Intel Software YouTube channel is a place to learn tips and tricks, get the latest news, watch product demos from both Intel, and our many partners across multiple fields. You'll find videos covering the topics listed below, and to learn more, you can follow the links provided!
Connect with Intel Software:
Visit INTEL SOFTWARE WEBSITE: https://intel.ly/2KeP1hD
Like INTEL SOFTWARE on FACEBOOK: http://bit.ly/2z8MPFF
Follow INTEL SOFTWARE on TWITTER: http://bit.ly/2zahGSn
INTEL SOFTWARE GITHUB: http://bit.ly/2zaih6z
INTEL DEVELOPER ZONE LINKEDIN: http://bit.ly/2z979qs
INTEL DEVELOPER ZONE INSTAGRAM: http://bit.ly/2z9Xsby
INTEL GAME DEV TWITCH: http://bit.ly/2BkNshu
#intelsoftware #openvino
Post Training Optimization - A Real Example of Integer Calibration | OpenVINO™ toolkit | Ep. 69