Skip to content

dpt.reshape changes strides of input array when reshaping to the same shape #1664

Closed
@antonwolfy

Description

@antonwolfy

The below example demonstrates different behavior in reshape method of NumPy and dpctl:

import numpy, dpctl, dpctl.tensor as dpt

dpctl.__version__
# Out: '0.17.0dev0+325.g0cb2181547'

a = dpt.ones((2, 1))
a.shape, a.strides
# Out: ((2, 1), (1, 1))

b = dpt.reshape(a, (2, 1))
b.shape, b.strides
# Out: ((2, 1), (1, 0))    # the output strides are changed

a = numpy.ones((2, 1)
a.shape, a.strides
# Out: ((2, 1), (8, 8))

b = numpy.reshape(a, (2, 1))
b.shape, b.strides
# Out: ((2, 1), (8, 8))    # the output strides are the same as for input array "a"

So, the question here if it's expected that the strides of output array might be changed when reshaping to the same shape?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions