numpy.require() function is useful for the array when correct flags is returned satisfies requirements for passing to compiled code (perhaps through ctypes).
Syntax: numpy.require(a, dtype=None, requirements=None)
Parameters:
a : array_like
dtype : data-type
requirements : str or list of str
The requirements list can be any of the following.
- ‘F’ : ‘F_CONTIGUOUS’ – ensure a Fortran-contiguous array.
- ‘C’ : ‘C_CONTIGUOUS’ – ensure a C-contiguous array.
- ‘A’ : ‘ALIGNED’ – ensure a data-type aligned array.
- ‘W’ : ‘WRITABLE’ – ensure a writable array.
- ‘O’ : ‘OWNDATA’ – ensure an array that owns its own data.
- ‘E’ : ‘ENSUREARRAY’ – ensure a base array, instead of a subclass.
Returns : ndarray
Exception : ValueError – Raises ValueError
Code #1:
Python3
# Python program explaining# numpy.require() function# importing numpyimport numpy as np# creating 4 x 4 arraydata = np.arange(16).reshape(4, 4)data.flags |
Output:
C_CONTIGUOUS : True F_CONTIGUOUS : False OWNDATA : False WRITABLE : True ALIGNED : True WRITEBACKIFCOPY : False UPDATEIFCOPY : False
Code #2:
Python3
import numpy as np# Python program explaining# numpy.require()b = np.require(data, dtype=np.float32, requirements=['A', 'W', 'O', 'C'])b.flags |
Output:
C_CONTIGUOUS : True F_CONTIGUOUS : False OWNDATA : True WRITABLE : True ALIGNED : True WRITEBACKIFCOPY : False UPDATEIFCOPY : False
