Python numpy array integer indexed flat slice assignment

Question

Was experimenting with numpy and found this strange behavior. This code works ok:

>>> a = np.array([[1, 2, 3], [4, 5, 6]])    
>>> a[:, 1].flat[:] = np.array([-1, -1])
>>> a 
array([[ 1, -1,  3],
       [ 4, -1,  6]])

But why this code doesn't change to -1 elements of 0 and 2 column?

>>> a[:, [0, 2]].flat[:] = np.array([-1, -1])
>>> a 
array([[ 1, -1,  3],
       [ 4, -1,  6]])

And how to write the code so that would change to -1 elements of 0 and 2 columns like this?

UPD: use of flat or smt similar is necessarily in my example

UPD2: I made example in question basing on this code:

img = imread(img_name)
xor_mask = np.zeros_like(img, dtype=np.bool)
# msg_bits looks like array([ True, False, False, ..., False, False,  True], dtype=bool)
xor_mask[:, :, channel].flat[:len(msg_bits)] = np.ones_like(msg_bits, dtype=np.bool)

And after assignment to xor mask with channel == 0 or 1 or 2 code works ok, but if channel == [1,2] or smt like this, assignment does not happen

It's not clear why you need to use flat. As already answered below, flat may create a copy therefore your updates may not change the original array. Explain why you need flat at all and maybe you can get a solution. — lbolla
– lbolla, Commented Oct 12, 2016 at 10:14
@ЮраМахоткин Seems you are stuck in an XY problem. — Divakar
– Divakar, Commented Oct 12, 2016 at 11:00
@Divakar yeah, you right. I will post question again but with correct problem formulation — KgOfHedgehogs
– KgOfHedgehogs, Commented Oct 12, 2016 at 11:04

Kasravnd · Accepted Answer · 2016-10-12 09:59:12Z

2

In first example by flattening the slice you don't change the shape and actually the ~~python~~ Numpy doesn't create a new object. so assigning to flattened slice is like assigning to actual slice. But by flattening a 2d array you're changing the shape and hence numpy makes a copy of it.

also you don't need to flatten your slice to add to it:

In [5]: a[:, [0, 2]] += 100

In [6]: a
Out[6]: 
array([[101,   2, 103],
       [104,   5, 106]])

edited Oct 12, 2016 at 9:59

answered Oct 12, 2016 at 9:40

Kasravnd

108k19 gold badges167 silver badges195 bronze badges

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

2 Comments

KgOfHedgehogs Over a year ago

This example from other problem, there is necessarily to use flat. Can you explain how to do with flat?

Kasravnd Over a year ago

@ЮраМахоткин Pleas add the use case to question, it's hard to answer like this.

lbolla · Accepted Answer · 2016-10-12 10:21:40Z

1

As others has pointed out .flat may create a copy of the original vector, so any updates to it would be lost. But flattening a 1D slice is fine, so you can use a for loop to update multiple indexes.

import numpy as np

a = np.array([[1, 2, 3], [4, 5, 6]])
a[:, 1].flat = np.array([-1, -1])
print a

# Use for loop to avoid copies
for idx in [0, 2]:
     a[:, idx].flat = np.array([-1, -1])
print a

Note that you don't need to use flat[:]: just flat is enough (and probably more efficient).

answered Oct 12, 2016 at 10:21

lbolla

5,4311 gold badge26 silver badges35 bronze badges

6 Comments

lbolla Over a year ago

What's not to like about it?

KgOfHedgehogs Over a year ago

Nothing, but i'm learning numpy and expected some numpy construction for this situation

lbolla Over a year ago

Well, you seem to be fixed on using flat, but flat must create a copy in your situation, so it can't be used. I think you simply have to "tune" your expectations...

KgOfHedgehogs Over a year ago

But wait. For-loop doesn't fit to my needs: xor_mask[:, :, channel].flat[:len(msg_bits)] = np.ones_like(msg_bits, dtype=np.bool) and for ch in channel: xor_mask[:, :, ch].flat[:len(msg_bits)] = np.ones_like(msg_bits, dtype=np.bool) logically does different process

lbolla Over a year ago

You've got to break msg_bits accordingly. I don't know why you are fixated with using flat, so I can't give you a meaningful recommendation.

|

berna1111 · Accepted Answer · 2016-10-12 10:02:41Z

You could just remove the flat[:] from a[:, [0, 2]].flat[:] += 100:

>>> import numpy as np
>>> a = np.array([[1, 2, 3], [4, 5, 6]])
>>> a[:, 1].flat[:] += 100
>>> a
array([[  1, 102,   3],
       [  4, 105,   6]])
>>> a[:, [0, 2]] += 100
>>> a
array([[101, 102, 103],
       [104, 105, 106]])

But you say it is necessary... Can't you just reshape whatever you are trying to add to the initial array instead of using flat?

The second index call makes a copy of the array while the first returns a reference to it:

>>> import numpy as np
>>> a = np.array([[1, 2, 3], [4, 5, 6]])
>>> b = a[:,1].flat
>>> b[0] += 100
>>> a
array([[  1, 102,   3],
       [  4,   5,   6]])
>>> b =a[:,[0,2]].flat
>>> b[0]
1
>>> b[0] += 100
>>> a
array([[  1, 102,   3],
       [  4,   5,   6]])
>>> b[:]
array([101,   3,   4,   6])

It appears that when the elements you wish to iterate upon in a flat maner are not adjacent numpy makes an iterator over a copy of the array.

Collectives™ on Stack Overflow

Python numpy array integer indexed flat slice assignment

3 Answers 3

2 Comments

6 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

6 Comments

Comments

Your Answer

Sign up or log in

Post as a guest