mygrad.Tensor.base

property Tensor.base: Optional[Tensor]

A reference to the base tensor that the present tensor is a view of.

It this tensor owns its memory, then this returns None.

Examples

The base of a tensor that owns its memory is None:

>>> import mygrad as mg
>>> x = mg.arange(5)
>>> x.base is None
True

Slicing creates a view, whose memory is shared with x:

>>> y = x[2:]
>>> y.base is x
True
>>> y.data.base is x.data
True

A view of a view has the same base as its “parent”

>>> z = y[:]
>>> z.base is x
True

The behavior of Tensor.base departs from that of ndarray.base in that mygrad will never create an “internal” tensor to serve as a base; e.g.

>>> import numpy as np
>>> np.reshape(2., (1,)).base
array(2.)

>>> mg.reshape(2., (1,)).base is None
True