a minimum Scala implementation for serving the fastText models
ft-serving is available on Maven Central.
libraryDependencies += "com.github.mskimm" %% "ft-serving" % "0.0.1"
$ sbt console
scala> val model = fasttext.FastText.load("data/cooking.ftz")
scala> model.predict("are egg whites generally available at the store ?", 1)
- supervised model (incl. quantized with
-qnorm=false -qout=false) with softmax loss
