- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi guys,
I am trying to convert the custom efficient- det model to onnx and to change as IR (openvino) format. I am able to convert the model to onnx. But when converting to onnx to openvino, I get the error as follows
"Check 'PartialShape(shape).refines(get_partial_shape())' failed at core/src/runtime/host_tensor.cpp:141:
Allocation shape {1} must be compatible with the partial shape: {0}"
Note: I am attaching a compressed file so you may reproduce this issue on your end (I will appreciate it!).
Thank you and will be waiting for your reply,
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi yaswanthan,
For your information, you can directly load ONNX model into OpenVINO™ toolkit Inference Engine without converting into Intermediate Representation (IR).
I was able to convert your ONNX model into Intermediate Representation (IR) with OpenVINO™ toolkit 2022.1.0. However, when inferencing the IR with Benchmark App, I got the same errors as you. Hence, I proceeded to infer ONNX model directly with Benchmark App but also getting the same errors.
Please have a check on your ONNX model to validate that the ONNX model itself is working fine.
Regards,
Peh
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi yaswanthan,
For your information, you can directly load ONNX model into OpenVINO™ toolkit Inference Engine without converting into Intermediate Representation (IR).
I was able to convert your ONNX model into Intermediate Representation (IR) with OpenVINO™ toolkit 2022.1.0. However, when inferencing the IR with Benchmark App, I got the same errors as you. Hence, I proceeded to infer ONNX model directly with Benchmark App but also getting the same errors.
Please have a check on your ONNX model to validate that the ONNX model itself is working fine.
Regards,
Peh
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi yaswanthan,
Thank you for your question. If you need any additional information from Intel, please submit a new question as this thread is no longer being monitored.
Regards,
Peh
![](/skins/images/B7BA37F9CE4B52FF7CAE1FB0388839C8/responsive_peak/images/icon_anonymous_message.png)
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page