Discrete Cosine Transform (DCT)

  Discrete Cosine Transform将一个在时间域(time domain)内长度为N的实值序列转换为另外一个
在频率域(frequency domain)内的长度为N的实值序列。下面是程序调用的例子。

  1. import org.apache.spark.ml.feature.DCT
  2. import org.apache.spark.ml.linalg.Vectors
  4. val data = Seq(
  5.   Vectors.dense(0.0, 1.0, -2.0, 3.0),
  6.   Vectors.dense(-1.0, 2.0, 4.0, -7.0),
  7.   Vectors.dense(14.0, -2.0, -5.0, 1.0))
  9. val df = spark.createDataFrame(data.map(Tuple1.apply)).toDF("features")
  11. val dct = new DCT()
  12.   .setInputCol("features")
  13.   .setOutputCol("featuresDCT")
  14.   .setInverse(false)
  16. val dctDf = dct.transform(df)
  17. dctDf.select("featuresDCT").show(3)