Module ss8 

ESM3 SS8 secondary-structure tokenization.

Token layout: 0 = PAD, 1 = MASK, 2 = UNK 3..10 = ‘G’,‘H’,‘I’,‘T’,‘E’,‘B’,‘S’,‘C’ (DSSP 8-class)

Functions

tokenize_ss8

Encode an SS8 secondary-structure string to token IDs.