Closed morgunl2 closed 4 years ago
This possibly an internal error in model optimizer (see #19). You can try to use OpenVINO scripts to download and convert model or try to convert it with previous OpenVINO release.
As @ZlodeiBaal suggested, fix it with roll back to torch 1.4 version (from 1.5).
Thank you for your help. I'll check it out.
Hope, it works.
Hey, so I am having the same issue. How do I install rollback to the previous issue 1.4 Pytorch?
Hi, if you are using pip
, then execute the following command:
pip install torch==1.4.0 torchvision==0.4.0 -f https://download.pytorch.org/whl/torch_stable.html
If not, select proper way from the official docs.
Same problem for me here. I have install 1.4 torch but it's not work
Hi! Use model downloader to download already converted model in OpenVINO format, if this does not work for some chance.
Hi! Use model downloader to download already converted model in OpenVINO format, if this does not work for some chance.
i 've done it already to but still error maybe because of my openvino version?
This is likely the case. But the model downloader downloads model already in OpenVINO format, so you cannot have such error (just nothing to convert).
This is likely the case. But the model downloader downloads model already in OpenVINO format, so you cannot have such error (just nothing to convert). i've found my error. It's because of my onnx, i create a new one and run it well. Thank you!
This is likely the case. But the model downloader downloads model already in OpenVINO format, so you cannot have such error (just nothing to convert).
I tried this approach, but the downloader only provides the pytorch checkpoint, no OpenVino model is downloaded using these scripts. Furthermore, their conversion script has the same issue as this repo. For anyone reading, just go straight for PyTorch downgrade.
Hi, if you are using
pip
, then execute the following command:pip install torch==1.4.0 torchvision==0.4.0 -f https://download.pytorch.org/whl/torch_stable.html
If not, select proper way from the official docs.
thanks! it works!
The resulting onnx file seems normal to me. Then, according to your manual:
Returns an error:
What can it be? Following the link provided, question 51 doesn't clear the situation. Thank you in advance!