Numpy structured array document confusion

Question

I am a total newbie to numpy structured arrays.

If I use the code from the docs (see here: Numpy Structured Array)

import numpy as np 
x=np.array([(1,2.,'Hello'), (2,3.,"World")], dtype=[('foo', 'i4'),('bar', 'f4'), ('baz', 'S10')])
x
y = x['foo']
y

x is correct: array([(1, 2.0, 'Hello'), (2, 3.0, 'World')], dtype=[('foo', '<i4'), ('bar', '<f4'), ('baz', 'S10')])

However y gives me array([1, 2]) and the docs say it should be array([ 2., 3.], dtype=float32)

I have a hard time believing the docs are wrong but this code is so short and and I copied/pasted it into python.

Am I doing something wrong?

Looks like a documentation error to me.

user2357112
– user2357112

2015-12-23 17:32:35 +00:00
Commented Dec 23, 2015 at 17:32 — user2357112
– user2357112, Commented Dec 23, 2015 at 17:32

Daniel · Accepted Answer · 2015-12-23 17:32:14Z

2

x['foo'] gives you an array of all first elements of each structure, so [1,2] is correct. The documentation mistakenly switched foo and bar, so read the docs as

y = x['bar']

and the rest of the example is correct.

answered Dec 23, 2015 at 17:32

Daniel

42.9k4 gold badges57 silver badges82 bronze badges

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

3 Comments

user918967 Over a year ago

Kinda crazy a noob can find this... How does someone go about correcting the docs?

ali_m Over a year ago

@user918967 You should open a new issue on the numpy GitHub page

hpaulj Over a year ago

docs.scipy.org/doc/numpy/user/index.html warns that the 'user guide' is a work in progress. Evidently developers have put more effort in to the reference section, as well as responding to code errors.

Collectives™ on Stack Overflow

Numpy structured array document confusion

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related