search

quentin-burthier / DCA

PyTorch implementation of Deep Communicating Agents for Abstractive Summarization
MIT License
8 stars 2 forks source link

why using word2vec rather than trained glove vectors as in paper #2

Open zide05 opened 4 years ago

zide05 commented 4 years ago

why using self trained word2vec to initialize the embedding rather than using trained glove vectors as paper discribed?