A question about random embedding at models/base.py.

nouhadziri / THRED

The implementation of the paper "Augmenting Neural Response Generation with Context-Aware Topical Attention"

https://arxiv.org/abs/1811.01063

MIT License

111 stars 25 forks source link

Closed shark-3 closed 5 years ago

shark-3 commented 5 years ago

Its shape is [vocab_size*embedding_size] in your code. Why isn't [vocab_size, embedding_size]?

nouhadziri commented 5 years ago

Yes you're right. We are working on fixing the problem and updating the code. Thanks.

ehsk commented 5 years ago

Fixed