Github user MLnick commented on a diff in the pull request: https://github.com/apache/spark/pull/11812#discussion_r56619995 --- Diff: mllib/src/main/scala/org/apache/spark/mllib/feature/Word2Vec.scala --- @@ -532,28 +539,14 @@ class Word2VecModel private[spark] ( blas.sgemv( "T", vectorSize, numWords, alpha, wordVectors, vectorSize, fVector, 1, beta, cosineVec, 1) - // Need not divide with the norm of the given vector since it is constant. val cosVec = cosineVec.map(_.toDouble) - var ind = 0 - val vecNorm = blas.snrm2(vectorSize, fVector, 1) --- End diff -- Removing this means that passing in a vector that is not in the vocab will not give the expected result as it is not explicitly normalized. As an example, some applications may pass in the average vector for a set of words to compute synonyms.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org