How to know what type of optimization have been done to my model when using OpenVINO model optimizer

[KSehairi]

Hello all,

I have optimized my Caffe model using the model optimizer from OpenVINO. I have set the precision calibration to 16.
Is there any method to know if the model optimizer has applied to my model layer fusion technique or model pruning. Also how to extract the memory performance?
Thanks in advance

[Aznie_Intel · Accepted solution]

Hi Ksehairi,

Which OpenVINO version you are using? You can look at this Model Optimization Techniques for the details of optimization operation in Model Optimizer. For the additional parameters when converting with Model Optimizer, you can check the list here and configure any option that works for you.

https://docs.openvino.ai/2024/notebooks/121-legacy-mo-convert-to-openvino-with-output.html#ir-preparation-with-python-conversion-api-and-model-optimizer-command-line-tool

You also may check and view your model layer using Netron to compare before and after converting.

Regards,

Aznie

[KSehairi]

Hi Aznie,
Thanks for your reply and these helpful links.
I am using the OpenVINO 2021.4.752, which I think is the latest that supports NCS2.

Best regards

[Aznie_Intel]

Hi Ksehairi,

This thread will no longer be monitored since this issue has been resolved. If you need any additional information from Intel, please submit a new question.

Regards,

Aznie