scale_to_z_score_per_key does not work with SparseTensors and key_vocabulary_filename is not None

[cyc]

If you try to run tft.scale_to_z_score_per_key while passing in SparseTensors for x and key, while also setting key_vocabulary_filename to a value other than None, you will get an error message like:

TypeError: Expected string, got <tensorflow.python.framework.sparse_tensor.SparseTensor object at 0x1a7d5e9a0> of type 'SparseTensor' instead.

This exception ultimately originates from here:

https://github.com/tensorflow/transform/blob/v1.3.0/tensorflow_transform/tf_utils.py#L649

The root cause is that key is a SparseTensor instead of a regular Tensor (key.values needs to be passed in instead of key)

[bharatjetti]

Hi @cyc 
Thank you for opening this issue. Since this issue has been open for a long time, the code/debug information for this issue may not be relevant with the current state of the code base.
The TensorFlow team is constantly improving the framework by fixing bugs and adding new features. We suggest you try the latest TensorFlow version with the latest compatible hardware configuration which could potentially resolve the issue. If you are still facing the issue, please create a new GitHub issue with your latest findings, with all the debugging information which could help us investigate.
Please follow the release notes to stay up to date with the latest developments which are happening in the TensorFlow space.”