pylingual/dev_scripts/statement

seq2seq

• train_tokenizer_auto.py:

  • trains the manual tokenizer

• tokenize_seq2seq.py:

  • tokenize the dataset for the seq2seq model

• train_seq2seq.py:

  • finetuning the pretrained model
  • will create a sequence-to-sequence translation model

• StatementConfiguration.py

  • defines the JSON format for statement translation training

manual1

Contains JSONs mapping bytecode instructions and their configurations to use in training.