Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

int8 PTQ using random

Luo__Andy
Novice
463 Views

Hi, 

  I want to do some quick perf benchmark for int8 quantized models but it is not time consuming to find and prepare all the calibration datasets. The accuracy is not a concern.

  Is there any way to generate int8 PTQ models with random input data? I did check the NNCF documentation and not find out how to do it. 

 

Regards,

Andy

0 Kudos
2 Replies
Aznie_Intel
Moderator
418 Views

Hi Luo_Andy,

 

Thanks for reaching out.

 

In general, you may use a random input dataset to apply 8-bit post-training quantization. NNCF API has two main capabilities, Basic quantization and Quantization with accuracy control where you need to prepare calibration datasets and validation datasets on your own if you are using random datasets. The minimum dataset should be 300 samples.

 

You may refer to this Dataset Preparation for Object Detection Sample to prepare your dataset.

 

 

Regards,

Aznie


0 Kudos
Aznie_Intel
Moderator
376 Views

Hi Luo_Andy,


This thread will no longer be monitored since we have provided information. If you need any additional information from Intel, please submit a new question. 



Regards,

Aznie


0 Kudos
Reply