Closed sadhoss closed 2 years ago
I have a follow up question to one of your additional observations:
For the "example" folder problem, did you add another folder or did you modify the code? I commented the section out but now it just prints out
It does not give any errors, but it does not create a tmp file.
I am on MacOS for context
@akdaly I just added the "example" folder. I Have not had an issue with it since.
Regarding the <stoi error when running compile_and_test.sh|ps1>, it has been a while since I worked on this, but I did/have gone past this error. One of those errors that just went away after a while. Will add a better explanation/resolution if I remember it!
@sadhoss Thank you! That could be really helpful, especially if you used MacOS, but if you do not remember no worries!
Describe the bug Following the PyTorch from Scratch Tutorial I get the following error, when trying to run detection through the mock-eye-module:
The model used is pulled from openvino workbench zoo. The default ssd model to test the mock-eye-module container. Hence error should not be related to the model
ssd_mobilenet_v2_coco
.Mock-eye-module by itself is compiling fine.
Based on simple debugging, error seems to be from the G-API when executed.
To Reproduce Steps to reproduce the behavior: Follow all steps in Prerequisites.
When converted model, and video person-bicycle-car-detection.mp4 downloaded.
I run the following on mac:
I run the following on windows:
Both return:
Expected behavior Based on the documentation, a window should pop out and stream a video with object detection overplayed.
Additional context Additional observation The scripts compile_and_test.sh expects an "example" folder in the mock-eye-module, not included when cloning the repo. The "mock-eye-module-debug" pulls "openvino/ubuntu18_runtime:latest" instead of "openvino/ubuntu18_runtime:2021.1".