Hi! For running Dataflow jobs it is sometimes required for us to ensure that Dataflow workers have certain required packages installed. For this reason, it would be useful to be able to determine the tensorflow_metadata version at runtime.
Ideally, this would follow the common python convention of a __version__ in the top-level module. For example, how tensorflow-data-validation does:
In [1]: import tensorflow_data_validation
In [2]: tensorflow_data_validation.__version__
Out[2]: '0.11.0'
Hi! For running Dataflow jobs it is sometimes required for us to ensure that Dataflow workers have certain required packages installed. For this reason, it would be useful to be able to determine the
tensorflow_metadata
version at runtime.Ideally, this would follow the common python convention of a
__version__
in the top-level module. For example, howtensorflow-data-validation
does: