IBM/MAX-Object-Detector

Docker image cannot be deployed in RedHat OpenShift due to permission error

ptitzler opened this issue · 0 comments

File "app.py", line 18, in <module>
    from api import ModelMetadataAPI, ModelLabelsAPI, ModelPredictAPI
    File "/workspace/api/__init__.py", line 18, in <module>
    from .predict import ModelLabelsAPI, ModelPredictAPI  # noqa
    File "/workspace/api/predict.py", line 35, in <module>
    model_wrapper = ModelWrapper()
    File "/workspace/core/model.py", line 45, in __init__
    serialized_graph = fid.read()
    File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/lib/io/file_io.py", line 125, in read
    self._preread_check()
    File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/lib/io/file_io.py", line 85, in _preread_check
    compat.as_bytes(self.__name), 1024 * 512, status)
    File "/opt/conda/lib/python3.6/site-packages/tensorflow/python/framework/errors_impl.py", line 528, in __exit__
    c_api.TF_GetCode(self.status.status))
    tensorflow.python.framework.errors_impl.PermissionDeniedError: assets/frozen_inference_graph.pb; Permission denied