Two sample parsers of ProteinNet records are provided, one for the text-based file format (text_parser.py) and one for the TFRecord based file format (tf_parser.py).
Text-based parser (text_parser.py)
python text_parser.py text_sample > out
This will print the python dict corresponding to each record in text_sample to standard output. The function read_record can also be used to process ProteinNet records programmatically.
TF-based parser (tf_parser.py)
This parser can only be used programmatically by invoking the read_protein function. For an example of its use see the RGN repository.