getting error to save the Matrix in txt file using numpy numpy

Question

I have written this code to make an array=>

import numpy as np
a = np.array([[[1,3],[1,4]],[[1,4],[6,9]]])
np.savetxt('test.txt',a,fmt='%d')

and getting this error =>

Traceback (most recent call last): File "/usr/local/lib/python3.4/dist-packages/numpy-1.11.2-py3.4-linux-x86_64.egg/numpy/lib/npyio.py", line 1158, in savetxt fh.write(asbytes(format % tuple(row) + newline)) TypeError: %d format: a number is required, not numpy.ndarray

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "", line 1, in File "/usr/local/lib/python3.4/dist-packages/numpy-1.11.2-py3.4-linux-x86_64.egg/numpy/lib/npyio.py", line 1162, in savetxt % (str(X.dtype), format)) TypeError: Mismatch between array dtype ('int64') and format specifier ('%d %d')

How to save the array as an integer in file using numpy?

what kind of structure are you expecting for your output file? can you post what you want? — Astrom, Apr 14 '17 at 15:30
you have a 3d array, try save `a.reshape(-1, 2)` instead. `np.savetxt('test.txt',a.reshape(-1,2),fmt='%d')` — Psidom, Apr 14 '17 at 15:36

score 2 · Answer 1 · answered Apr 14 '17 at 15:36

2

I am not sure what kind of output you want...

You can do this

F = open('test.txt', 'w')
for i in a:
    for j in i:
        line = '%s \t %s \n'%(j[0], j[1])
        F.write(line)

F.close()

Will give you

1 3

1 4

6 9

answered Apr 14 '17 at 15:36

Astrom

735
7
19

if I want to load this matrix How I will do this?? – Sudip Das Apr 14 '17 at 15:40
then you can use `numpy.genfromtxt('test.txt')` – Astrom Apr 14 '17 at 15:40
please not that the comment from @Psidom will give you the same as my proposition in a more _numpythonic_ way – Astrom Apr 14 '17 at 15:41
yes but you have posted that one is quite difficult to load the matrix from file. but if I use that one, is quite good cause there have a function loadtxt() in numpy. then why I will use this one – Sudip Das Apr 14 '17 at 15:45
for the difference between `numpy.genfromtxt()` and `numpy.loadtxt()` you can look [there](http://stackoverflow.com/questions/20245593/difference-between-numpy-genfromtxt-and-numpy-loadtxt-and-unpack) – Astrom Apr 14 '17 at 15:51

score 1 · Answer 2 · answered Apr 14 '17 at 15:30

%d format: a number is required, not numpy.ndarray

Essentially you can't use a format string on an object/class/what-have-you; its expecting a number, not an array of numbers

convert the array to a printable string. Similarly, you can't really save integers in files, only strings in files.

What you'll want to do is take each row in the array and write it to the file, something similar to this:

to_write = ''
for small_list in list_of_lists:
    to_write += ''.join(small_list, ',') + "\n"
my_file.write(to_write)

if I want to load this matrix How I will do this?? – Sudip Das Apr 14 '17 at 15:40 — Sudip Das, Apr 14 '17 at 15:40

score 1 · Accepted Answer · edited May 23 '17 at 12:34

There is an old thread on this problem

Saving numpy array to csv produces TypeError Mismatch

If you check, your array is a 3D array which is causing the issue

1. If you translate it to a 2D it will work

import numpy as np a = np.array([[[1,3],[1,4]],[[1,4],[6,9]]]) np.savetxt('test.txt',a.reshape(-1,a.shape[-1]),fmt="%d")

2. If you write one row at a time it will work too

But this will have issues reading back

My suggestion, write it as dict/json and convert it to numpy after reading

getting error to save the Matrix in txt file using numpy numpy

3 Answers3