This is a project for CS 388 Natural Language Processing, spring 2021 semester. In this project, I created a semantic parser using an Sequence2Sequence model built in PyTorch to perform semantic parsing on a geoquery dataset. The LSTM based encoder was provided by the course. The evaluation code was developed by Robert Jia and Percy Liang of Stanford.
Create a decoder module in PyTorch, combine it with the encoder, perform training, and perform inference. I had to implement two different Decoder modules: a basic LSTM based decoder, and an LSTM decoder with an additional attention layer.
Two examples from the dataset are provided below:
what is the population of atlanta ga ?
_answer ( A , ( _population ( B , A ) , _const ( B , _cityid ( atlanta , _ ) ) ) )
what states border texas ?
_answer ( A , ( _state ( A ) , _next_to ( A , B ) , _const ( B , _stateid ( texas ) ) ) )