So I’d like to get better at understanding model design and implementation. I was wondering if anyone had tips on reading research papers and figuring out how to implement the models in them.
For example I’d like to dive into understanding the QANet (https://arxiv.org/abs/1804.09541) model for machine comprehension.
In this case, there’s a nice write-up provided Min Sang Kim (https://medium.com/@minsangkim/implementing-question-answering-networks-with-cnns-5ae5f08e312b) that implements it in Tensorflow. I was planning on seeing if I can replicate the approach in Pytorch.
But in cases where there isn’t a tutorial or example code, any tips would be super helpful!