Sahar/samples ovep escalation #8
Conversation
| ## Prerequisites | ||
| 1. [The Intel<sup>®</sup> Distribution of OpenVINO toolkit](https://docs.openvinotoolkit.org/latest/index.html) | ||
| 1. [The Intel<sup>®</sup> Distribution of OpenVINO toolkit](https://docs.openvino.ai/latest/openvino_docs_install_guides_install_runtime.html) | ||
| Please select Install OpenVINO Runtime using an installer |
There was a problem hiding this comment.
can we also add the direct hyperlink to this page.
https://docs.openvino.ai/latest/openvino_docs_install_guides_installing_openvino_linux.html#doxid-openvino-docs-install-guides-installing-openvino-linux
| ## Running the ONNXRuntime OpenVINO Execution Provider sample | ||
| ```bash | ||
| python3 tiny_yolov2_obj_detection_sample.py | ||
| python3 tiny_yolov2_obj_detection_sample.py --video bottle-detection.mp4 --model tinyyolov2.onnx |
There was a problem hiding this comment.
Can we also make passing device type options configurable with the sample.
Default, it should pick OpenVINO-EP CPU.
But the user should also have options to run on other device_types in OpenVINO-EP and also MLAS (default CPU-EP)
|
|
||
|
|
||
| ## Running the ONNXRuntime OpenVINO Execution Provider sample | ||
| ```bash |
There was a problem hiding this comment.
Also we need to add --h option to this sample
|
Is this comment addressed "There is a new error related to API version mismatch with numpy. (see screenshots)" |
|
Is there a way, the user will be notified that his sample is falling back to MLAS even though he tried to run it on OpenVINO-EP. |
|
Licenses need to be updated for both the samples. change from: Copyright (C) 2021, Intel Corporation |
…o sahar/samples_ovep_escalation
…tel/onnxruntime-inference-examples into sahar/samples_ovep_escalation
We will have to put some change in code to know if we are doing subgraph partitioning, it is currently not there |
Draft PR to update the samples