How to change numpy array by removing quotations and adding a comma between each value?

Question

If I have an array like this:

 array([['10      2       1967    19.7    7.5'],
        ['10      3       1967    18.3    21.0'],
        ['10      4       1967    17.6    0.0']])

How do I remove the quotations and add a comma in between each value? Such that I get something like this:

 array([[10, 2, 1967, 19.7, 7.5],
        [10, 3, 1967, 18.3, 21.0],
        [10, 4, 1967, 17.6, 0.0]])

Do you want to convert the single numbers to something? int or float? With a numpy array you can't mix both. — a_guest
– a_guest, Commented Jan 28, 2019 at 16:45
What the source of this array? It would be better to correct it before it's created, rather fix it after. The source and targets are very different arrays. One is 1d with string elements. The other is 2d with numeric elements. The quotes and commas tell us about the arrays, but they aren't what define them. — hpaulj
– hpaulj, Commented Jan 28, 2019 at 16:56
I missed something before. Your source array is 2d, with shape (3,1). I neglected that inner set of []. The key descriptors of a numpy array are shape and dtype. — hpaulj
– hpaulj, Commented Jan 28, 2019 at 18:23

jpp · Accepted Answer · 2019-01-28 16:45:52Z

2

Regular NumPy arrays must have a single dtype. In this case, float may be appropriate:

A = np.array([['10      2       1967    19.7    7.5'],
              ['10      3       1967    18.3    21.0'],
              ['10      4       1967    17.6    0.0']])

B = np.array([x[0].split() for x in A], dtype=float)

# array([[   10. ,     2. ,  1967. ,    19.7,     7.5],
#        [   10. ,     3. ,  1967. ,    18.3,    21. ],
#        [   10. ,     4. ,  1967. ,    17.6,     0. ]])

answered Jan 28, 2019 at 16:45

jpp

166k37 gold badges301 silver badges362 bronze badges

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

Comments

Rakesh · Accepted Answer · 2019-01-28 16:48:00Z

0

This is one approach.

from numpy import array
import ast
l = array([['10      2       1967    19.7    7.5'],
        ['10      3       1967    18.3    21.0'],
        ['10      4       1967    17.6    0.0']])

l = [list(map(ast.literal_eval,  j.split()))  for i in l for j in i]
print(l)

Output:

[[10, 2, 1967, 19.7, 7.5], [10, 3, 1967, 18.3, 21.0], [10, 4, 1967, 17.6, 0.0]]

Note: You cannot have both int and float in np.array.

answered Jan 28, 2019 at 16:48

Rakesh

82.9k17 gold badges85 silver badges122 bronze badges

Comments

iGian · Accepted Answer · 2019-01-28 17:59:11Z

0

Other couple of options:

array = np.array([['10      2       1967    19.7    7.5'],
                  ['10      3       1967    18.3    21.0'],
                  ['10      4       1967    17.6    0.0']])

new_ary1 = np.array([ [ float(n) for n in e[0].split() ] for e in array ])
new_ary2 = np.array([ np.array(e[0].split(), dtype=float) for e in array ])

answered Jan 28, 2019 at 17:59

iGian

11.2k3 gold badges24 silver badges38 bronze badges

Collectives™ on Stack Overflow

How to change numpy array by removing quotations and adding a comma between each value?

3 Answers 3

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related