vehicle-detection-0202 responds to queries when running on CPU, gives an error when running on GPU

[lugi0]

I am using the model vehicle-detection-0202 in its openvino_ir format running in a OVMS environment, however when I try to send requests to the model running on a CPU I can get back the expected response, e.g.

{
    "model_name": "my-model",
    "model_version": "1",
    "outputs": [{
            "name": "detection_out",
            "shape": [1, 1, 200, 7],
            "datatype": "FP32",
            "data": [0.0, ..., data, ..., 0.0]
        }]
}

If I instead try to send a request to the model running on a GPU I keep getting error messages like the following:

{"code":2,"message":"json: unsupported value: -Inf"}

(Sometimes instead of -Inf the value is NaN). I don't quite understand what I'm doing wrong, as the requests are in the same format and the data has all the same attributes. Any idea what's going on?

[Wovchena]

Try running benchmark_app -d GPU -m vehicle-detection-0202.xml using the same OpenVINO version as in OVMS environment. This issue should be transferred to https://github.com/openvinotoolkit/openvino if it fails. If it passes, it's https://github.com/openvinotoolkit/model_server failure. I'll pass it to them for now. By the way, please, specify the hardware and your OVMS environment.

[Wovchena]

I can't transfer it. It will take some time to figure that out. Meanwhile you can submit a duplicate to one of them depending on benchmark_app test.