Ch-6 Store as TF Records

[vishwanath79]

For example on creating TFRecords - Consider adding the following functions to store the packed BytesList and Int64List as per the documentation to assist the reader, throws an error otherwise.

def _int64_feature(value):
  return tf.train.Feature(int64_list=tf.train.Int64List(value=[value]))

def _bytes_feature(value):
  return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value]))

[sidgan]

Thank you for bringing this up. Addressed via #105.