Tutorial installed different torch packages a couple of days apart.

[wb666greene]

Using Convert and Optimize YOLOv8 real-time object detection with OpenVINO™ as a guide, I've setup two Ubuntu 22.04 systems
following the tutorial about maybe a week apart and got these versions installed last week from the command :
pip install "torch>=2.1" "torchvision>=0.16" "ultralytics==8.2.24" onnx tqdm opencv-python --extra-index-url https://download.pytorch.org/whl/cpu:
torch 2.3.1+cpu
torchvision 0.18.1+cpu

And got these versions after doing it yesterday:
torch 2.4.0+cpu
torchvision 0.19.0+cpu

So far things seem to be working, so no real issue to report yet, but wouldn't it be wise to lock these installed versions to what was specified when the tutorial was created, instead of just assuming that newer versions will be "backward compatible" and not break anything? Few things are as frustrating when starting out as having weird issues when trying to learn from a tutorial or sample code because one of the underlying pieces has changed.

I do have one intermittent core dump:
[INFO] Using OpenVINO: 2024.2.0-15519-5c0f38f83f6-releases/2024/2 clang: CommandLine Error: Option 'h' registered more than once! LLVM ERROR: inconsistency in registered CommandLine options Aborted (core dumped)
But this is a known Ubuntu 22.04 issue so nothing OpenVINO can fix.

[wb666greene]

Edit: I just realized these discussions cover all the tutorials, I'm specifically using code lifted from this tutorial:
https://github.com/openvinotoolkit/openvino_notebooks/blob/latest/notebooks/yolov8-optimization/yolov8-object-detection.ipynb

I don't think these version differences mattered, as I created a new VENV environment and used the == option for the pip install to match the versions I got on a different system a week ago and it made no difference, still get the issue.

[INFO] Using OpenVINO: 2024.2.0-15519-5c0f38f83f6-releases/2024/2
Available devices:
   CPU: Intel(R) Celeron(R) N4020 CPU @ 1.10GHz
   GPU: Intel(R) UHD Graphics 600 [0x3185] (iGPU)
Traceback (most recent call last):
  File "/home/ai/LearnTestOV/yolo8ov.py", line 45, in <module>
    det_compiled_model = core.compile_model(det_ov_model, "GPU", ov_config)
  File "/home/ai/xyz/lib/python3.10/site-packages/openvino/runtime/ie_api.py", line 543, in compile_model
    super().compile_model(model, device_name, {} if config is None else config),
RuntimeError: Exception from src/inference/src/cpp/core.cpp:104:
Exception from src/inference/src/dev/plugin.cpp:54:
Check 'false' failed at src/plugins/intel_gpu/src/plugin/program_builder.cpp:185:
[GPU] ProgramBuilder build failed!
Exception from src/plugins/intel_gpu/src/runtime/ocl/ocl_stream.cpp:429:
[GPU] clWaitForEvents, error code: -14

I been running yolo4 ( https://github.com/TNTWEN/OpenVINO-YOLOV4 ) on this iGPU for over a year when the system was running Ubuntu 20.04 with openvino_2021.3.394 so I didn't expect problems with GPU compatibility this time..

Here are the "specs" of the GPU in question:

An older system "Whiskey-Lake" with UHD 620 works fine with this code on Ubuntu22.04
I also have 22.04 on a similar Celeron N5105 with UHD graphics(JSL) that ran this tutorial code fine.