search

wlin12 / wang2vec

Extension of the original word2vec using different architectures
Apache License 2.0
210 stars 49 forks source link

weightedword2vec #3

Open wangshaonan opened 8 years ago

wangshaonan commented 8 years ago

Hi Wang, Thanks for sharing the code. I have the following two questions. What can weightedword2vec do ? Is there the implementation of attention based cbow model ?

wlin12 commented 8 years ago

Hi,

So, the weightedword2vec code basically expects each sentence to have a weight (e.g., 0.12) as the first word, which will be multiplied by the gradient of the predictor function (ne, nce or hs).

I haven’t uploaded the code for the attention-based cbow model yet.

Cheers, Wang Ling

On Feb 29, 2016, at 8:55 AM, wang shaonan notifications@github.com wrote:

Hi Wang, Thanks for sharing the code. I have the following two questions. What can weightedword2vec do ? Is there the implementation of attention based cbow model ?

— Reply to this email directly or view it on GitHub https://github.com/wlin12/wang2vec/issues/3.

wangshaonan commented 8 years ago

Got it. Thank you.