Package

org.apache.flink.ml

math

Permalink

package math

Convenience methods to handle Flink's org.apache.flink.ml.math.Matrix and Vector abstraction.

Linear Supertypes
AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. math
  2. AnyRef
  3. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Type Members

  1. trait BreezeVectorConverter[T <: Vector] extends Serializable

    Permalink

    Type class which allows the conversion from Breeze vectors to Flink vectors

    Type class which allows the conversion from Breeze vectors to Flink vectors

    T

    Resulting type of the conversion, subtype of Vector

  2. case class DenseMatrix(numRows: Int, numCols: Int, data: Array[Double]) extends Matrix with Serializable with Product

    Permalink

    Dense matrix implementation of Matrix.

    Dense matrix implementation of Matrix. Stores data in column major order in a continuous double array.

    numRows

    Number of rows

    numCols

    Number of columns

    data

    Array of matrix elements in column major order

  3. case class DenseVector(data: Array[Double]) extends Vector with Serializable with Product

    Permalink

    Dense vector implementation of Vector.

    Dense vector implementation of Vector. The data is represented in a continuous array of doubles.

    data

    Array of doubles to store the vector elements

  4. trait Matrix extends AnyRef

    Permalink

    Base trait for a matrix representation

  5. implicit class RichMatrix extends Iterable[(Int, Int, Double)]

    Permalink
  6. implicit class RichVector extends Iterable[(Int, Double)]

    Permalink
  7. class SparseMatrix extends Matrix with Serializable

    Permalink

    Sparse matrix using the compressed sparse column (CSC) representation.

    Sparse matrix using the compressed sparse column (CSC) representation.

    More details concerning the compressed sparse column (CSC) representation can be found [http://en.wikipedia.org/wiki/Sparse_matrix#Compressed_sparse_column_.28CSC_or_CCS.29].

  8. case class SparseVector(size: Int, indices: Array[Int], data: Array[Double]) extends Vector with Serializable with Product

    Permalink

    Sparse vector implementation storing the data in two arrays.

    Sparse vector implementation storing the data in two arrays. One index contains the sorted indices of the non-zero vector entries and the other the corresponding vector entries

  9. trait Vector extends Serializable

    Permalink

    Base trait for Vectors

  10. trait VectorBuilder[T <: Vector] extends Serializable

    Permalink

    Type class to allow the vector construction from different data types

    Type class to allow the vector construction from different data types

    T

    Subtype of Vector

Value Members

  1. object BLAS extends Serializable

    Permalink

    BLAS routines for vectors and matrices.

    BLAS routines for vectors and matrices.

    Original code from the Apache Spark project: http://git.io/vfZUe

  2. object Breeze

    Permalink

    This class contains convenience function to wrap a matrix/vector into a breeze matrix/vector and to unwrap it again.

  3. object DenseMatrix extends Serializable

    Permalink
  4. object DenseVector extends Serializable

    Permalink
  5. object SparseMatrix extends Serializable

    Permalink
  6. object SparseVector extends Serializable

    Permalink
  7. object Vector extends Serializable

    Permalink
  8. object VectorBuilder extends Serializable

    Permalink
  9. package distributed

    Permalink
  10. def vector2Array(vector: Vector): Array[Double]

    Permalink

    Stores the vector values in a dense array

    Stores the vector values in a dense array

    vector

    Subtype of Vector

    returns

    Array containing the vector values

Inherited from AnyRef

Inherited from Any

Ungrouped