Proper data type support for updated version of TF Lite

[phildow]

I'm seeing a number of these errors:

There was a problem writing the data buffer to the tensor, error: Error Domain=org.tensorflow.lite.interpreter Code=1 "Input tensor at index (0) expects data size (8), but got (4)." UserInfo={NSLocalizedDescription=Input tensor at index (0) expects data size (8), but got (4).}

Hypothesis 1: Trying to write int32s to int64 inputs. Are int64s supported in the updated version of TF Lite?

[phildow]

We are adding support for int32 and int64, which are also supported by our full TF models.