Use the following to run different models from src directory:
python run.py [--h] [--t training_file] [-e testing_file] [-m model]
List of models can be found:
- 0: LSTM with Character CNN for attribute classification
- 1: Manhattan Siamese LSTM with Character CNN for attribute similarity
- 2: Hierarchical Attention Network for attribute classification
- 3: TFIDF (should be an attribute similarity method)
Embedding pretrained models can be downloaded from: https://nlp.stanford.edu/projects/glove/
Pickled data files can be downloaded from: https://drive.google.com/file/d/1Hs1hCf4MVy38Kt8ecFCCaK1c_1amEopZ/view?usp=sharing