matrix-vector multiplication
matrix-vector multiplication
Structure-optimized mmul
Hadamard product
Hadamard product
in-place Hadamard product.
in-place Hadamard product. We probably don't want to use assign to optimize for sparse operations, in case of Hadamard product it really can be done
5.
5.0 *=: A is equivalent to A *= 5.0. Included for completeness.
A *=: B is equivalent to B *= A.
A *=: B is equivalent to B *= A. Included for completeness.
A := B - A which is -(A - B)
A := B - A which is -(A - B)
1.
1.0 /: A is eqivalent to R's 1.0/A
Elementwise deletion
1.
1.0 /=: A is equivalent to A = 1.0/A in R
Sparse function assign: iterate and assign over non-zeros only
Sparse function assign: iterate and assign over non-zeros only
Sparse assign: iterate and assign over non-zeros only
Sparse assign: iterate and assign over non-zeros only
Functional assign with (Double) => Double
Functional assign with (Double) => Double
Assigning from a row-wise collection of vectors
Ideally, we would probably want to override equals().
Ideally, we would probably want to override equals(). But that is not possible without modifying AbstractMatrix implementation in Mahout which would require discussion at Mahout team.
test if rank == min(nrow,ncol).
Warning: This provides read-only view only.
Warning: This provides read-only view only.
In most cases that's what one wants. To get a copy,
use m.t cloned
transposed view
Warning: this creates a clone (as in mx * -1), in many applications inplace inversion mx *= -1
might be an infinitely better choice.
Warning: this creates a clone (as in mx * -1), in many applications inplace inversion mx *= -1
might be an infinitely better choice.