search

hengruo / QANet-pytorch

A PyTorch implementation of QANet.
MIT License
344 stars 67 forks source link

Some difference between the paper and the model #11

Closed wlhgtc closed 6 years ago

wlhgtc commented 6 years ago
2018-07-17 8 37 41

According to the original paper, the dimension of input to Encoder block is 500(200+300). That means we should not change the dimension of word embedding and char embedding in "Embedding Layer" . We should use "Conv" at the start of "Encoder Layer" in order to map it to dimension.

hengruo commented 6 years ago

You are right. I realized that issue but adopted the current design because it seems more compact... I merged your code.