convert the lines of a file into numpy array

Question

I have a files in which data is something like this

[1,2,3,4,5]
[4,2,3,4,5,6]
[3,12,3,34,5]
[3,4,5,6,7,8]
[2,3,4,5,6,6]
[2,3,4,5,5,4,3]
[1,2,3,23,2,1]

i just want to convert this into numpy array, something like this

[[1,2,3,4,5],[4,2,3,4,5,6]]

I tried following code:

import ast
import numpy as np
a = np.array([])
with open("test.txt","r") as f:
        for line in f:
                line = ast.literal_eval(line)
                np.append(line,a)

it is only giving []

I tried following code: It might prove useful if you also were to divulge what happened next. — Paul Panzer
– Paul Panzer, Commented Feb 16, 2018 at 6:03
Btw. I think you want to swap arguments of append and more importantly reassign to a. np.append does not work in-place. — Paul Panzer
– Paul Panzer, Commented Feb 16, 2018 at 6:15
BTW, this isn't going to be a proper array, since the dimensions aren't regular. — juanpa.arrivillaga
– juanpa.arrivillaga, Commented Feb 16, 2018 at 6:22

Rakesh · Accepted Answer · 2018-02-16 07:02:02Z

1

Using array method in numpy.

import ast
from numpy import array

p = "PATH_TO_INPUT_FILE"
data = []
with open(p,"r") as f:
    for i in f.readlines():
        data.append(ast.literal_eval(i.strip()))
data = array(data)
print(data)
print(type(data))

Output:

[[1, 2, 3, 4, 5] [4, 2, 3, 4, 5, 6] [3, 12, 3, 34, 5] [3, 4, 5, 6, 7, 8]
 [2, 3, 4, 5, 6, 6] [2, 3, 4, 5, 5, 4, 3] [1, 2, 3, 23, 2, 1]]
<type 'numpy.ndarray'>

answered Feb 16, 2018 at 7:02

Rakesh

82.9k17 gold badges86 silver badges122 bronze badges

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

Comments

Osman Mamun · Accepted Answer · 2018-02-16 06:18:20Z

0

with open('file.txt') as f:
    data = []
    for line in f:
        words = line.replace('[', ''). replace(']', '').replace('\n', '').split(',')
        data.append([int(i) for i in words])

if you want numpy array:

with open('file.txt') as f:
    data = []
    for line in f:
        words = line.replace('[', ''). replace(']', '').replace('\n', '').split(',')
        data.append(np.array([int(i) for i in words]))

answered Feb 16, 2018 at 6:18

Osman Mamun

2,8802 gold badges18 silver badges24 bronze badges

3 Comments

ggupta Over a year ago

it is giving: [array([1, 2, 3, 4, 5]), array([4, 2, 3, 4, 5, 6]), array([ 3, 12, 3, 34, 5]), array([3, 4, 5, 6, 7, 8]), array([2, 3, 4, 5, 6, 6]), array([2, 3, 4, 5, 5, 4, 3]), array([ 1, 2, 3, 23, 2, 1])]

Osman Mamun Over a year ago

The first one will give you a list of lists (which is what you wanted in your question). The second one will produce list of np arrays. Use np.array(data) to the first output to get 2d np array

Mr. T Over a year ago

@ggupta What do you expect from numpy? Your lists don't have the same dimension, how are they supposed to be converted into an n x m array?

Collectives™ on Stack Overflow

convert the lines of a file into numpy array

2 Answers 2

Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related