Clojure: using Spark factory method with Java interop

Question

I think this is a Java interop question, but perhaps it's a nuance of Spark (or failing to import something from Scala?). I am trying to follow along with the example here . It includes the following code:

import org.apache.spark.mllib.linalg.Vector;
import org.apache.spark.mllib.linalg.Vectors;
Vector dv = Vectors.dense(1.0, 0.0, 3.0);

My first thought was this:

(import '[org.apache.spark.mllib.linalg Vector Vectors])
(Vectors/dense 1.0 0.0 3.0)

But I get:

CompilerException java.lang.IllegalArgumentException: No matching method: dense, compiling:(/tmp/form-init6598386874684927469.clj:1:1)

I tried:

(.dense (Vectors. 1.0 0.0 3.0))

and got:

CompilerException java.lang.IllegalArgumentException: No matching ctor found for class org.apache.spark.mllib.linalg.Vectors, compiling:(/tmp/form-init6598386874684927469.clj:1:9)

I've tried a variety of additional combinations (vectors, Java arrays, the .. macro, etc.), but can't get it to work. In particular, even though the documentation shows a constructor that takes no arguments ( see here ), nothing like (Vectors.) seems to work (a No matching ctor ) error. Thanks.

Answer 1

Did you try (Vectors/dense (double-array [1.0 0.0 3.0])) ? I didn't test this myself but it should match the Vector dense (double[] values) method in Vectors .