Numpy saves Integer Array as Float Array

Question

My array of Integers gets saved as an array of Floats.

array = [1, 2, 3, 4, 5, 6, 7];
print(array)
>> [1, 2, 3, 4, 5, 6, 7]

np.savetxt('labels.txt', array, delimiter=",");

test = np.loadtxt("labels.txt", delimiter=",")
print(test)
>> [1. 2. 3. 4. 5. 6. 7.]

My created labels.txt looks like this:

1.000000000000000000e+00 2.000000000000000000e+00 3.000000000000000000e+00 4.000000000000000000e+00 5.000000000000000000e+00 6.000000000000000000e+00 7.000000000000000000e+00

But I expect the text to be 1, 2, 3, 4, 5, 6, 7. How can I achieve this?

The default save format is: fmt='%.18e'. The savetxt docs discusses altenative fmt values. — hpaulj
– hpaulj, Commented Jan 3, 2020 at 20:56
You can control the format that you save the array using fmt of savetxt, then you can also specify the dtype during loadtxt when you read back in. — Cory Kramer
– Cory Kramer, Commented Jan 3, 2020 at 20:57
@prune, I am removing the duplicate. That SO is about assigning elements in a np.empty array. This is about writing values to a csv. — hpaulj
– hpaulj, Commented Jan 3, 2020 at 23:28

alexpiers · Accepted Answer · 2020-01-03 21:07:50Z

2

Use the fmt input of %i (integer formatting) to np.savetxt

import numpy as np

testList = [1, 2, 3, 4, 5]
np.savetxt("testList.txt", [testList], fmt="%i", delimiter=",", newline="\n")

This gives the output 1,2,3,4,5. Note that the input to savetxt is a list of lists. That is because the delimiter separates column values and the newline separates rows (so you need 2D to get the delimiters).

answered Jan 3, 2020 at 21:07

alexpiers

7265 silver badges12 bronze badges

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

Comments

Gulzar · Accepted Answer · 2020-01-03 21:08:27Z

1

If you don't have anything confining you to np.savetext, then try using pickle:

import numpy as np
import pickle as pkl

path = r"path/to/dump.pkl"
a = np.array([1,2,3,4,5,6,7])
b = pkl.dump(a, path)
c = pkl.load(path)
np.testing.assertequal(c, a)

Or in console if you feel more like it,

>>> import numpy as np
>>> a = [1, 2, 3, 4, 5, 6, 7]
>>> a
[1, 2, 3, 4, 5, 6, 7]
>>> b = np.array(a)
>>> b
array([1, 2, 3, 4, 5, 6, 7])
>>> import pickle
>>> pickle.dumps(b)
b'\x80\x03cnumpy.core.multiarray\n_reconstruct\nq\x00cnumpy\nndarray\nq\x01K\x00\x85q\x02C\x01bq\x03\x87q\x04Rq\x05(K\x01K\x07\x85q\x06cnumpy\ndtype\nq\x07X\x02\x00\x00\x00i4q\x08K\x00K\x01\x87q\tRq\n(K\x03X\x01\x00\x00\x00<q\x0bNNNJ\xff\xff\xff\xffJ\xff\xff\xff\xffK\x00tq\x0cb\x89C\x1c\x01\x00\x00\x00\x02\x00\x00\x00\x03\x00\x00\x00\x04\x00\x00\x00\x05\x00\x00\x00\x06\x00\x00\x00\x07\x00\x00\x00q\rtq\x0eb.'
>>> p = pickle.dumps(b)
>>> d = pickle.loads(p)
>>> d
array([1, 2, 3, 4, 5, 6, 7])

This won't be a human readable file, but it answers your requirement.

or, just

with open(path, 'wb') as f:
    np.savetxt(f, a, fmt='%i', delimiter=",")

Which I personally favor less than the standard pickle

edited Jan 3, 2020 at 21:08

answered Jan 3, 2020 at 20:58

Gulzar

28.7k42 gold badges158 silver badges260 bronze badges

2 Comments

Jonas Over a year ago

I'm not required to use np but I need it to be a .txt file

Gulzar Over a year ago

@Jonas Also added the csv option.

Collectives™ on Stack Overflow

Numpy saves Integer Array as Float Array

2 Answers 2

Comments

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related