Implement some deep model for Text Matching
DIIN
Natural Language Inference over Interaction Space
ICLR2018
ESIM
Enhanced LSTM for Natural Language Inference
ACL2017
MatchPyramid
Text Matching as Image Recognition
AAAI16
Easy to train. You just need to edit the code in preprocessor.py to load your data
and just run the command python -m textmatch.train --classfier=DIIN to train a diin model.
Easy to extend. if you want to explore a new model. just to create a new file in textmatch/model
and define a class in this file, which the class extends the TextModel in the current dir. you just to override method get_model.
last just register in the textmatch/model/__init__.py