Hi,

How about this?

import org.apache.spark.ml.linalg._
val toSV = udf((v: Vector) => v.toDense)
val df = Seq((0.1, Vectors.sparse(16, Array(0, 3), Array(0.1, 0.3))), (0.2, Vectors.sparse(16, Array(0, 3), Array(0.1, 0.3)))).toDF("a", "b")
df.select(toSV($"b"))

// maropu


On Mon, Nov 14, 2016 at 1:20 PM, janardhan shetty <janardhanp22@gmail.com> wrote:
Hi,

Is there any easy way of converting a dataframe column from SparseVector to DenseVector  using

import org.apache.spark.ml.linalg.DenseVector API ?

Spark ML 2.0



--
---
Takeshi Yamamuro