snlp_term_project_group10

Encoder Decoder

Instructions to run the code:

  1. python3 reading_pairs.py
  2. python3 preprocessing.py
  3. python3 train.py

Adaptor Grammar

For training: python -m launch_train --input_directory=./poetry/ --output_directory=./ --grammar_file=./poetry/grammar.unigram --number_of_documents=18274 --batch_size=10

For testing: python -m launch_test --input_directory=./poetry/ --model_directory=./poetry/ --non_terminal_symbol=Word