TurnkeyML shows an unhelpful error message when docker daemon is not running

[jeremyfowers]

Issue

The message below is shown when the docker daemon is not running

Models discovered during profiling:

resnet50.py:
        model (executed 1x)
                Model Type:     Pytorch (torch.nn.Module)
                Class:          ResNet (<class 'timm.models.resnet.ResNet'>)
                Location:       C:\\Users\\danie\\miniconda3\\envs\\tkml\\lib\\site-packages\\timm\\models\\_builder.py, line 390
                Parameters:     25,557,032 (97.49 MB)
                Input Shape:    'Positional Arg 1': (1, 3, 224, 224)
                Hash:           694a8fff
                Build dir:      C:\Users\danie/.cache/turnkey/resnet50_timm_694a8fff
                Status:         Unknown turnkey error: 'Total Latency'
                Traceback (most recent call last):
                  File "C:\Users\danie\onnxmodelzoo\toolchain\src\turnkeyml\analyze\script.py", line 193, in explore_invocation
                    perf = benchmark_model(
                  File "C:\Users\danie\onnxmodelzoo\toolchain\src\turnkeyml\model_api.py", line 109, in benchmark_model
                    perf = model_handle.benchmark()
                  File "C:\Users\danie\onnxmodelzoo\toolchain\src\turnkeyml\run\basert.py", line 177, in benchmark
                    mean_latency=self.mean_latency,
                  File "C:\Users\danie\onnxmodelzoo\toolchain\src\turnkeyml\run\tensorrt\runtime.py", line 54, in mean_latency
                    return float(self._get_stat("Total Latency")["mean "].split(" ")[1])
                  File "C:\Users\danie\onnxmodelzoo\toolchain\src\turnkeyml\run\basert.py", line 190, in _get_stat
                    return performance[stat]
                KeyError: 'Total Latency'

Suggested solution

The errors shown below are suppressed when TURNKEY_DEBUG is not set to True

docker: error during connect: this error may indicate that the docker daemon is not running: Post "http://%2F%2F.%2Fpipe%2Fdocker_engine/v1.24/containers/create?name=tensorrt23.03-py3": open //./pipe/docker_engine: The system cannot find the file specified.
See 'docker run --help'.
error during connect: this error may indicate that the docker daemon is not running: Get "http://%2F%2F.%2Fpipe%2Fdocker_engine/v1.24/containers/tensorrt23.03-py3/json": open //./pipe/docker_engine: The system cannot find the file specified.
error during connect: this error may indicate that the docker daemon is not running: Post "http://%2F%2F.%2Fpipe%2Fdocker_engine/v1.24/containers/tensorrt23.03-py3/stop": open //./pipe/docker_engine: The system cannot find the file specified.

The ideal solution here is to show those errors as part of the Status

Reproducing

On a Windows system run:
turnkey models\timm\resnet50.py --device nvidia