TensorFlow’s High Level API uses SavedModel (see: https://www.tensorflow.org/versions/r1.0/api_docs/python/tf/saved_model) in the export_savedmodel() function (see: https://www.tensorflow.org/api_docs/python/tf/estimator/Estimator#export_savedmodel). When I understand its descriptions correctly, this enables a model to be saved (=frozen?) in a language neutral way.

SavedModel might be an alternative to your approch, which seems to be the expected format for e.g. Google Cloud ML.

Unfortunately, its documentation is a bit confusing and complex, which all its SignatureDefs etc.