module Common: sig
.. end
Types and functions common to all precision dependent sub-modules.
type
trans2 = [ `N | `T ]
Transpose parameter (normal or transposed)
type
side = [ `L | `R ]
Side parameter (left or right)
type
diag = [ `N | `U ]
Diagonal parameter (unit or non-unit)
type
norm2 = [ `I | `O ]
Type of 1-norm (`O
) and infinity norm (`I
)
type
norm4 = [ `F | `I | `M | `O ]
Type of 1-norm (`O
), infinity norm (`I
) and the Frobenius norm (`F
).
`M
is the maximum of the absolute values (not a true matrix norm).
type
svd_job = [ `A | `N | `O | `S ]
SVD computation flags
exception InternalError of string
InternalError msg
gets raised when BLAS or LAPACK exhibit undefined
behaviour.
type
int_vec = (int, Bigarray.int_elt, Bigarray.fortran_layout) Bigarray.Array1.t
Type of OCaml integer vectors.
val create_int_vec : int -> int_vec
create_int_vec n
Returns an int-vector with n
rows.
type
int32_vec = (int32, Bigarray.int32_elt, Bigarray.fortran_layout) Bigarray.Array1.t
Type of 32bit Fortran integer vectors.
val create_int32_vec : int -> int32_vec
create_int32_vec n
Returns an int32-vector with n
rows.
val mat_from_vec : ('a, 'b, 'c) Bigarray.Array1.t -> ('a, 'b, 'c) Bigarray.Array2.t
mat_from_vec a
converts the vector a
into a matrix with Array1.dim a
rows and 1 column. The data is shared between the two matrices.