generate the response for out of domain query
- QA_generation.py file help to generate question answer pair save in pickle format
- generate_data.py generate dataset where pass parameter -t=train/test and -n=number of QA pair want
- word_tokernized.py help its create source and target vocabulary and clean the data
- reader.py used for prepare data for model
- try_sqs_with_attension.py create seq2seqmodel where decoder use multiple LSTMCell(either GRUCell) for encoder and dencoder. Also,Apply attension and beam search for decoding
- model.py used simple encoder decoder method with beam search