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.

Error Executing Ported Model in OpenVino

A__Siva
Beginner
770 Views

#1. Trained a Pytorch Model

#2. Converted it into ONNX format. Conversion to ONNX format is successful No Errors

#3. Used OpenVino Option to get the XML using Model Optimizer. The Model Optimizer Generates Required XML No Errors. All the layers mentioned in XML are Supported by OpenVino. 

#4. Got this error while running the demo - [ERROR] MKLDNNGraph::CreateGraph: such topology cannot be compiled for dynamic batch.

With the XML this error cannot be compiled for dynamic batch returned 

Any thoughts / suggestion to take it forward ? 

 

0 Kudos
1 Reply
Shubha_R_Intel
Employee
770 Views

Dear Siva, 

My guess is that your ONNX model has a layer which is not supported by Dynamic Batching. Or maybe you're trying to run this on MYRIAD - which is a problem since Dynamic Batching is only supported on CPU and GPU. Can you kindly check this documentation to see how its content may apply to your situation ? I'm not sure which demo you are running, but does it set KEY_DYN_BATCH_ENABLED as described in the code below?

https://docs.openvinotoolkit.org/latest/_docs_IE_DG_DynamicBatching.html

Thanks,

Shubha

0 Kudos
Reply