Save numpy array to CSV without scientific notation

Question

I was able to calculate the mean, min and max of A:

 import numpy as np

 A = ['33.33', '33.33', '33.33', '33.37']

 NA = np.asarray(A)
 NA = NA.astype(float)
 AVG = np.mean(NA, axis=0)
 MN = np.min(NA, axis=0)
 MX = np.max(NA, axis=0)

 print AVG, MN, MX

What is the easiest way to save those results to a csv documento using python? I dont have one so it needs to be created.

If I use this:

 np.savetxt('datasave.csv', (AVG,MN,MX), delimiter=',')

It will show as scientific notation in csv. How do I not get that but float?

Answer 1

In [153]: print(AVG, MN, MX)
33.34 33.33 33.37

The default write:

In [154]: np.savetxt('test.txt',(AVG, MN, MX), delimiter=',')
In [155]: cat test.txt
3.333999999999999631e+01
3.332999999999999829e+01
3.336999999999999744e+01

write with a custom fmt:

In [156]: np.savetxt('test.txt',(AVG, MN, MX), delimiter=',', fmt='%f')
In [157]: cat test.txt
33.340000
33.330000
33.370000

Or if you want values on one line, make it a 2d array (or equivalent with extra []), so it writes one row per line:

In [160]: np.savetxt('test.txt',[[AVG, MN, MX]], delimiter=',', fmt='%f')
In [161]: cat test.txt
33.340000,33.330000,33.370000

There are other parameters that you can experiment with.

Answer 2

You can use the CSV module

import csv
f = open('New file.csv',"wb")
writer = csv.writer(f)
for row in [AVG, MN, MX]:
    writer.writerow(row)

f.close()