Skip to content

Latest commit

 

History

History
24 lines (20 loc) · 825 Bytes

README.md

File metadata and controls

24 lines (20 loc) · 825 Bytes

Please refer to ProjectDescription.pdf and ProjectPresentation.pdf for a full report of this project.

qclass_dl

Question classification on TREC dataset (and its Vietnamese translation) using Deep Neural structures, namely CNN & LSTM & both of them.

Much of the code is reused from https://github.com/dennybritz/cnn-text-classification-tf https://github.com/yoonkim/CNN_sentence https://github.com/tensorflow/tensorflow/tree/master/tensorflow/models/rnn/ptb https://github.com/tensorflow/tensorflow/tree/master/tensorflow/models/rnn/translate

To see list of parameters

./train_cnn.py --help
./train_cnnlstm.py --help
./train_lstm.py --help

To run

./train_cnn.py --vn
./train_cnnlstm.py --lstm_type='gru'
./train_lstm.py --batch_size=100