python: how to convert list of tuples to numpy array

Question

I have list of tuples (x,y coordinates) converted to a string and written in a file. When I read line from file, my list looks like this:

[(341, 115), (174, 227), (457, 308)]

how do I convert this kind of list to numpy array? end result should look like this:

[[341 115]
 [174 227]
 [457 308]]

np.array(literal_eval("[(341, 115), (174, 227), (457, 308)]")) ? — Enayat
– Enayat, Commented Mar 12, 2019 at 12:56

DirtyBit · Accepted Answer · 2019-03-12 13:34:56Z

3

Using numpy:

lst = [(341, 115), (174, 227), (457, 308)]

import numpy as np
print(np.array(lst))

OUTPUT:

[[341 115]
 [174 227]
 [457 308]]

Using list comprehension:

print([list(lst) for lst in lst])

OUTPUT:

[[341, 115], [174, 227], [457, 308]]

EDIT:

If it is a string which it does not look like in the code pasted in the question:

lst = "[(341, 115), (174, 227), (457, 308)]"

Then:

import numpy as np
from ast import literal_eval
print(np.array(literal_eval(lst)))

Eventually (for list comprehension):

print([list(lst) for lst in literal_eval(lst)])

answered Mar 12, 2019 at 12:54

DirtyBit

16.8k5 gold badges37 silver badges56 bronze badges

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

1 Comment

[int(x) for x in lst] is redundant as all elements in lst are already of type int. Just use list(lst).

meowgoesthedog · Accepted Answer · 2019-03-12 12:51:56Z

1

Use ast.literal_eval to convert the line (a string) to an actual list object, and call numpy.array directly on it:

line = "[(341, 115), (174, 227), (457, 308)]"

from ast import literal_eval
import numpy as np

np.array(literal_eval(line))

Output:

array([[341, 115],
   [174, 227],
   [457, 308]])

answered Mar 12, 2019 at 12:51

meowgoesthedog

15.1k4 gold badges30 silver badges43 bronze badges