verry.linalg.IntervalMatrix#

class verry.linalg.IntervalMatrix(inf, sup=None, **kwargs)#

Bases: ABC, Generic

Abstract base class for inf-sup type interval matrices.

Parameters:

  • inf (ndarray | Sequence)

  • sup (ndarray | Sequence, optional)

Variables:

  • inf (ndarray) – Infimum of the interval matrix.

  • sup (ndarray) – Supremum of the interval matrix.

classmethod empty(shape)#

Return a new interval matrix of given shape, without initializing entries.

classmethod eye(n, m=None)#

Return a matrix with ones on the diagonal and zeros elsewhere.

classmethod ones(shape)#

Return a new interval matrix of given shape, filled with ones.

classmethod zeros(shape)#

Return a new interval matrix of given shape, filled with zeros.

copy()#

Return a copy of the interval matrix.

diam()#

Return component-wise upper bounds of the diameter.

empty_like()#

Return a new interval matrix with the same shape as self.

flatten()#

Return a copy of the interval matrix collapsed into one dimension.

interiorcontains(other)#

Return True if the interior of the interval matrix contains other.

isbounded()#

Return True if both inf and sup are bounded.

isdisjoint(other)#

Return True if the interval matrix has no elements in common with other.

issubset(other)#

Test whether every element in the interval matrix is in other.

issuperset(other)#

Test whether every element in other is in the interval matrix.

mid()#

Return an approximation of the midpoint.

As with scalar intervals, x.mid() in x is guaranteed to be True.

ones_like()#

Return an interval matrix of ones with the same shape as self.

rad()#

Return component-wise upper bounds of the radius.

reshape(shape)#

Give a new shape to an interval matrix without changing its data.

transpose()#

Return a view of the transposed matrix.

zeros_like()#

Return an interval matrix of zeros with the same shape as self.

property T#

Shorthand for transpose().

See also

transpose

property ndim#

Shorthand for len(self.shape).

Unlike NumPy’s ndarray, x.ndim is always greater than or equal to 1, i.e., there is no 0D matrix.

property shape#

Tuple of matrix dimensions.

Examples

>>> from verry.linalg import FloatIntervalMatrix as FIM
>>> x = FIM([[1, 2, 3], [4, 5, 6]])
>>> y = FIM([1.0, 2.0])
>>> x.shape
(2, 3)
>>> y.shape
(2,)