I was trying to get familiar with mxnet framework and was browsing the tutorial section - https://mxnet.apache.org/tutorials/index.html.
I did not come across any illustrations for generating vector representations of words, like word2vec models( https://en.wikipedia.org/wiki/Word2vec ). Would it make sense to create a PR with one such tutorial. Is that something that will make sense in this code repo.
@astonzhang
I think @zackchase is doing it.
Is there already a PR in review?
Will MxNet implement a tokenizer as the one in Keras? Right now I use Keras to process the raw data and then convert it numpy array. It works well but too stupid. 馃槥
@anirudhacharya If you are familiar with Keras, you can follow my data iter first.
comment toxic
@anirudhacharya you can find example of word embedding training and usage in the Gluon-NLP project:
Please consider closing, thanks
Most helpful comment
Will MxNet implement a tokenizer as the one in Keras? Right now I use Keras to process the raw data and then convert it numpy array. It works well but too stupid. 馃槥
@anirudhacharya If you are familiar with Keras, you can follow my data iter first.
comment toxic