Numeric

The numeric module provides methods that are lacking from the numpy module. An accompanying extension module _numeric.c should be compiled to benefit from extra performance, although a Python-only implementation is provided as fallback. A warning message is printed if the extension module is not found.

nutils.numeric.normdim(ndim, n)

check bounds and make positive

nutils.numeric.align(arr, trans, ndim)

create new array of ndim from arr with axes moved accordin to trans

nutils.numeric.get(arr, axis, item)

take single item from array axis

nutils.numeric.expand(arr, *shape)

nutils.numeric.linspace2d(start, stop, steps)

linspace & meshgrid combined

nutils.numeric.contract(A, B, axis=-1)

nutils.numeric.contract_fast(A, B, naxes)

contract last n axes

nutils.numeric.dot(A, B, axis=-1)

Transform axis of A by contraction with first axis of B and inserting remaining axes. Note: with default axis=-1 this leads to multiplication of vectors and matrices following linear algebra conventions.

nutils.numeric.fastrepeat(A, nrepeat, axis=-1)

repeat axis by 0stride

nutils.numeric.fastmeshgrid(X, Y)

mesh grid based on fastrepeat

nutils.numeric.meshgrid(*args)

multi-dimensional meshgrid generalisation

nutils.numeric.appendaxes(A, shape)

append axes by 0stride

nutils.numeric.inverse(A)

linearized inverse

nutils.numeric.determinant(A)

nutils.numeric.eig(A, sort=False)

Compute the eigenvalues and vectors of a hermitian matrix sort -1/0/1 -> descending / unsorted / ascending

nutils.numeric.eigh(A, sort=False)

Compute the eigenvalues and vectors of a hermitian matrix sort -1/0/1 -> descending / unsorted / ascending

nutils.numeric.reshape(A, *shape)

more useful reshape

nutils.numeric.mean(A, weights=None, axis=-1)

generalized mean

nutils.numeric.norm2(A, axis=-1)

L2 norm over specified axis

nutils.numeric.normalize(A, axis=-1)

devide by normal

nutils.numeric.cross(v1, v2, axis)

cross product

nutils.numeric.stack(arrays, axis=0)

powerful array stacker with singleton expansion

nutils.numeric.bringforward(arg, axis)

bring axis forward

nutils.numeric.diagonalize(arg)

append axis, place last axis on diagonal of self and new