Open shu0o0 opened 1 year ago
Hi @shu0o0yX, ONNX Sequence-based operations are currently not supported by OpenVINO. Some initial research work is in progress, but I am not able to say any specific date.
Hi @shu0o0yX, ONNX Sequence-based operations are currently not supported by OpenVINO. Some initial research work is in progress, but I am not able to say any specific date.
OK! Thanks for your reply.
hey @shu0o0yX, could you share a name/source of the model, if it is not a problem?
hey @shu0o0yX, could you share a name/source of the model, if it is not a problem?
OK! The link is https://github.com/OFA-Sys/DAFlow. Hope that helps.
it will, thank you!
Internal ticket to track enabling this model: 101224
Internal ticket to track enabling this model: 101224
What does that mean?
Internal ticket to track enabling this model: 101224
What does that mean?
This is an internal reference number to an internal issue-tracking system.
Is it a problem of my model? What should I do to solve it? Plateform is Windows 10, It is my openvino version: openvino 2022.3.0 openvino-dev 2022.3.0 openvino-telemetry 2022.3.0
It's error: [ ERROR ] Traceback (most recent call last): File "F:\Anaconda3\envs\pytorch\Lib\site-packages\openvino\tools\mo\main.py", line 50, in main ngraph_function = convert_model(argv) File "F:\Anaconda3\envs\pytorch\lib\site-packages\openvino\tools\mo\convert.py", line 47, in convert_model return _convert(args) File "F:\Anaconda3\envs\pytorch\lib\site-packages\openvino\tools\mo\convert_impl.py", line 937, in _convert raise e.with_traceback(None) RuntimeError: Check 'unknown_operators.empty()' failed at C:\Jenkins\workspace\private-ci\ie\build-windows-vs2019\b\repos\openvino\src\frontends\onnx\frontend\src\core\graph.cpp:215: OpenVINO does not support the following ONNX operations: ConcatFromSequence, SequenceEmpty, SplitToSequence
Are you also working on virtual try-on recently? I converted several models to OpenVINO IR Format but all failed.
Is it a problem of my model? What should I do to solve it? Plateform is Windows 10, It is my openvino version: openvino 2022.3.0 openvino-dev 2022.3.0 openvino-telemetry 2022.3.0 It's error: [ ERROR ] Traceback (most recent call last): File "F:\Anaconda3\envs\pytorch\Lib\site-packages\openvino\tools\mo\main.py", line 50, in main ngraph_function = convert_model(argv) File "F:\Anaconda3\envs\pytorch\lib\site-packages\openvino\tools\mo\convert.py", line 47, in convert_model return _convert(args) File "F:\Anaconda3\envs\pytorch\lib\site-packages\openvino\tools\mo\convert_impl.py", line 937, in _convert raise e.with_traceback(None) RuntimeError: Check 'unknown_operators.empty()' failed at C:\Jenkins\workspace\private-ci\ie\build-windows-vs2019\b\repos\openvino\src\frontends\onnx\frontend\src\core\graph.cpp:215: OpenVINO does not support the following ONNX operations: ConcatFromSequence, SequenceEmpty, SplitToSequence
Are you also working on virtual try-on recently? I converted several models to OpenVINO IR Format but all failed.
Maybe virtual try-on models are not fit for OpenVINO.
This issue will be closed in 2 weeks in case of no activity.
This issue will be closed in a week because of 9 months of no activity.
Is it a problem of my model? What should I do to solve it? Plateform is Windows 10, It is my openvino version: openvino 2022.3.0 openvino-dev 2022.3.0 openvino-telemetry 2022.3.0
It's error: [ ERROR ] Traceback (most recent call last): File "F:\Anaconda3\envs\pytorch\Lib\site-packages\openvino\tools\mo\main.py", line 50, in main ngraph_function = convert_model(argv) File "F:\Anaconda3\envs\pytorch\lib\site-packages\openvino\tools\mo\convert.py", line 47, in convert_model return _convert(args) File "F:\Anaconda3\envs\pytorch\lib\site-packages\openvino\tools\mo\convert_impl.py", line 937, in _convert raise e.with_traceback(None) RuntimeError: Check 'unknown_operators.empty()' failed at C:\Jenkins\workspace\private-ci\ie\build-windows-vs2019\b\repos\openvino\src\frontends\onnx\frontend\src\core\graph.cpp:215: OpenVINO does not support the following ONNX operations: ConcatFromSequence, SequenceEmpty, SplitToSequence