why is numpy.fromstring reading numbers wrong?

Question

I am writing code that uses numpy.fromstring to read arrays from xml element text.

It runs with no error, but what it reads is very strange.

for example

import numpy as np

nr = 24
r_string = '''
0.0000    0.0100    0.0200    0.0300    0.0400    0.0500    0.0600    0.0700
0.0800    0.0900    0.1000    0.1100    0.1200    0.1300    0.1400    0.1500
0.1600    0.1700    0.1800    0.1900    0.2000    0.2100    0.2200    0.2300
'''

r = np.fromstring(r_string, count = nr)

print(r)

prints the following(garbage)

[1.20737375e-153 1.48440234e-076 1.30354286e-076 6.96312257e-077
 6.01356142e-154 1.20737830e-153 1.82984908e-076 1.30354286e-076
 6.96312257e-077 3.22522589e-086 6.01347037e-154 6.03686893e-154
 1.39804459e-076 9.72377416e-072 3.24245662e-086 6.01347037e-154
 6.03686880e-154 1.39939399e-076 1.79371973e-052 1.91654811e-076
 8.54289848e-072 6.96312257e-077 6.01356142e-154 1.20738399e-153]

What is going on here?

I will appreciate help here.

By using the default sep, you trigger a deprecated use of the function. From the documentation: Passing sep='', the default, is deprecated since it will trigger the deprecated binary mode of this function. This mode interprets string as binary bytes, rather than ASCII text with decimal numbers. — Bram Vanroy
– Bram Vanroy, Commented Dec 17, 2019 at 12:46
See the docs: sep : str, optional The string separating numbers in the data; extra whitespace between elements is also ignored. — Fourier
– Fourier, Commented Dec 17, 2019 at 12:46

Dishin Goyani · Accepted Answer · 2019-12-17 12:44:44Z

2

you need to declare sep=' '

>>> r = np.fromstring(r_string, count = nr, sep=' ')
>>> r
array([0.  , 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1 ,
       0.11, 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 0.18, 0.19, 0.2 , 0.21,
       0.22, 0.23])

answered Dec 17, 2019 at 12:44

Dishin Goyani

7,7533 gold badges33 silver badges42 bronze badges

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

3 Comments

Deena Roller Over a year ago

Thank you, now it works. It is strange that 'sep' is defined as optional in the manual link

Dishin Goyani Over a year ago

As other also explain the default, is deprecated since it will trigger the deprecated binary mode of this function - since version 1.14 - you can also see more about it in document link you have already mentaion above

Dishin Goyani Over a year ago

@DeenaRoller Happy to help, and I see you're new to SO If this answer or any other one solved your issue, please mark it as accepted.

Fourier · Accepted Answer · 2019-12-17 14:05:23Z

0

Running np.fromstring without the sep specified will actually throw the warning:

DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs.

You need to specify your seperator, like:

np.fromstring(r_string, sep="\t")

Output:

array([0.  , 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1 ,
       0.11, 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 0.18, 0.19, 0.2 , 0.21,
       0.22, 0.23])

edited Dec 17, 2019 at 14:05

answered Dec 17, 2019 at 12:44

Fourier

3,0033 gold badges29 silver badges40 bronze badges

Collectives™ on Stack Overflow

why is numpy.fromstring reading numbers wrong?

2 Answers 2

3 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related