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

How to correctly convert ONNX models

Fde__Antoni
Beginner
137 Views

Hello,

I am trying to use inference engine with ONNX model. I converted inception_v2 and resnet50, both from links from https://software.intel.com/en-us/articles/OpenVINO-Using-ONNX#supported-topologies, using mo.py with input model and output dir. Then I loaded the models into classification_sample application but the predictions I am getting for input images are not even close to the correct categories, I tried few different images. Before I run caffe models (from model downloader) using the same application and the predictions were ok, so I think the problem might be with the conversion?

0 Kudos
0 Replies
Reply