lib

Type Members

1. case class DColWiseMatrix[Elem, T](data: DList[(Elem, Iterable[(Elem, T)])]) extends Product with Serializable

   

   A col-wise matrix.

   A col-wise matrix. This is an efficient representation for multiplying by an in-memory vector. The contents of each column must be small enough to fit in memory

2. case class DMatrix[Elem, Value](data: DList[((Elem, Elem), Value)])(implicit evidence$24: WireFormat[Elem], evidence$25: Ordering[Elem], evidence$26: WireFormat[Value]) extends Product with Serializable

   

   A distributed Matrix, stored in coordinate format.

   A distributed Matrix, stored in coordinate format. Operations are optimised assuming the matrix is both large and sparse, but shouldn't be too terrible for large dense matrixes.

3. case class DRowWiseMatrix[Elem, T](data: DList[(Elem, Iterable[(Elem, T)])])(implicit evidence$19: WireFormat[Elem], evidence$20: Ordering[Elem], evidence$21: WireFormat[T]) extends Product with Serializable

   

   A distributed row-wise matrix.

   A distributed row-wise matrix. This is an efficient representation for multiplying by an in-memory vector. The contents of each row must be small enough to fit in memory

4. case class DVector[Elem, T](data: DList[(Elem, T)])(implicit evidence$1: WireFormat[Elem], evidence$2: Ordering[Elem], evidence$3: WireFormat[T]) extends Product with Serializable

   

   A distributed vector, stored in coordinate form.

   A distributed vector, stored in coordinate form. Optimised for a huge spare vector, but should perform reasonably for a large dense vector too. If the vector is small (i.e. a few megabytes or less, you'll probably be better off with the in memory vectors instead

5. case class InMemDenseVector[T](data: DObject[IndexedSeq[T]])(implicit evidence$16: WireFormat[T]) extends Product with Serializable

   

   An efficient, dense vector that needs to be small enough to fit in memory.

   An efficient, dense vector that needs to be small enough to fit in memory. It must be indexed by an int

6. case class InMemVector[Elem, T](data: DObject[Map[Elem, T]])(implicit evidence$10: WireFormat[Elem], evidence$11: Ordering[Elem], evidence$12: WireFormat[T]) extends Product with Serializable

   

   A vector that is small enough to send to one mapper (i.e.

   A vector that is small enough to send to one mapper (i.e. a few megs or less). Best suited for a sparse vector

7. trait Library extends AnyRef

   

8. case class Relational[K, A](left: DList[(K, A)])(implicit evidence$1: WireFormat[K], evidence$2: Grouping[K], evidence$3: WireFormat[A]) extends Product with Serializable