flopscope.

flopscope.numpy.ix_

Construct an open mesh from multiple sequences.

Adapted from NumPy docs np.ix_

Areacore
Typecustom
NumPy Refnp.ix_
Cost
per-operation
Flopscope Context

Construct open mesh from multiple sequences. Cost: numel(output).

This function takes N 1-D sequences and returns N outputs with N dimensions each, such that the shape is 1 in all but one dimension and the dimension with the non-unit shape value cycles through all N dimensions.

Using ix_ one can quickly construct index arrays that will index the cross product. a[flops.ix_([1,3],[2,5])] returns the array [[a[1,2] a[1,5]], [a[3,2] a[3,5]]].

Parameters

args:1-D sequences

Each sequence should be of integer or boolean type. Boolean sequences will be interpreted as boolean masks for the corresponding dimension (equivalent to passing in flops.nonzero(boolean_sequence)).

Returns

out:tuple of ndarrays

N arrays with N dimensions each, with N the number of input sequences. Together these arrays form an open mesh.

See also

Examples

>>> import flopscope.numpy as fnp
>>> a = flops.arange(10).reshape(2, 5)
>>> a
array([[0, 1, 2, 3, 4],
       [5, 6, 7, 8, 9]])
>>> ixgrid = flops.ix_([0, 1], [2, 4])
>>> ixgrid
(array([[0],
       [1]]), array([[2, 4]]))
>>> ixgrid[0].shape, ixgrid[1].shape
((2, 1), (1, 2))
>>> a[ixgrid]
array([[2, 4],
       [7, 9]])
>>> ixgrid = flops.ix_([True, True], [2, 4])
>>> a[ixgrid]
array([[2, 4],
       [7, 9]])
>>> ixgrid = flops.ix_([True, True], [False, False, True, False, True])
>>> a[ixgrid]
array([[2, 4],
       [7, 9]])