GloVe has been unmaintained and broken for a long time. The only people asking about it where those who were stumped by the broken example. GloVe on its own has one intrinsic problem: a vocabulary of size N requires to build a matrix of NxN. This resulted in suboptimal performance for most people. Overall, there are better performing approaches these days so it isn't worth the effort to fix this implementation. Signed-off-by: Paul Dubs <paul.dubs@gmail.com> |
||
---|---|---|
.. | ||
deeplearning4j-nlp | ||
deeplearning4j-nlp-chinese | ||
deeplearning4j-nlp-japanese | ||
deeplearning4j-nlp-korean | ||
deeplearning4j-nlp-uima | ||
pom.xml |