What does the array flags attribute tell?

Question

In a programming quiz I came accross the flags attribute for numpy arrays in Python and I was wondering was the output all means. After reading https://docs.scipy.org/doc/numpy/reference/generated/numpy.ndarray.flags.html, I don't really understand some of the output. What do C_CONTIGUOUS, F_CONTIGUOUS, ALIGNED and UPDATEIFCOPY mean, explained without as much jargon as the documentation?

import numpy as np
x=np.array([[3,4],[3,5]])
print(x.flags)

which outputs the following:

  C_CONTIGUOUS : True
  F_CONTIGUOUS : False
  OWNDATA : True
  WRITEABLE : True
  ALIGNED : True
  UPDATEIFCOPY : False

tupui · Accepted Answer · 2017-07-29 19:09:08Z

1

It gives information about how the array is stored in memory.

C_CONTIGUOUS true means it is stored as c-type in memory.

See the documentation for more on each attribute.

answered Jul 29, 2017 at 19:09

tupui

6,5463 gold badges35 silver badges55 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

Femkemilene Over a year ago

Thank you, but I'm trying to figure out what kind of information. I understand OWNDATA and WRITABLE and starting to understand C_CONTIGUOUS and F_CONTIGUOUS as well (but not why in a higher dimensional array they can both be true).

tupui Over a year ago

@Femkemilene I guess it is because internally each dimension are stored separately. So on can be a c-type and the other a f-type.

Femkemilene Over a year ago

I don't understand your answer. As far as I understand the arrays are always stored contiguous (so not seperately?), for C-contiguous the rows are stored next to each other, for F-contiguous the columns.. Would I be right to assume that the only way in which higher-dimensional arrays can be stored in the same way is that they only have one element in each dimension?

tupui Over a year ago

@Femkemilene you are right data is contiguous as detailed here docs.scipy.org/doc/numpy/reference/arrays.ndarray.html

tupui Over a year ago

@Femkemilene it seems that it is only possible if self.shape[k]=1

Collectives™ on Stack Overflow

What does the array flags attribute tell?

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related