Numpy savetxt to csv with integer integer string

One problem is this:

c_(list1, list2, list3)

c_ is a quirky object that is used by indexing it instead of calling it. When you call it, you get the error mentioned in the comments:

In [42]: c_(list1, list2, list3)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-42-8fcbef857684> in <module>()
----> 1 c_(list1, list2, list3)

TypeError: 'CClass' object is not callable

Instead, use square brackets, so you are indexing:

In [44]: c_[list1, list2, list3]
Out[44]: 
array([['10', '15', 'foo'],
       ['20', '25', 'bar'],
       ['30', '35', 'baz']], 
      dtype='|S3')

Note that it creates an array with dtype '|S3'. That is, all the elements in the array have been converted to strings. To save this with savetxt, use fmt='%s, %s, %s':

In [45]: savetxt('result.csv', c_[list1, list2, list3], fmt='%s, %s, %s')

In [46]: !cat result.csv
10, 15, foo
20, 25, bar
30, 35, baz

Also, instead of c_[list1, list2, list3], you could use zip(list1, list2, list3). Then the savetxt function will handle converting that argument to an array.

In [57]: list1 = [100000, 200000, 300000]

In [58]: savetxt('result.csv', zip(list1, list2, list3), fmt='%s, %s, %s')

In [59]: !cat result.csv
100000, 15, foo
200000, 25, bar
300000, 35, baz

Apparently c_ doesn't do a good job figuring out the appropriate lengths of the strings:

In [60]: c_[list1, list2, list3]
Out[60]: 
array([['100', '15', 'foo'],
       ['200', '25', 'bar'],
       ['300', '35', 'baz']], 
      dtype='|S3')

By the way, your data is not already in a numpy array, so I don't think you gain much by using savetxt instead of the standard library csv. For example,

In [61]: import csv

In [62]: with open('result.csv', 'w') as f:
   ....:     wrtr = csv.writer(f)
   ....:     wrtr.writerows(zip(list1, list2, list3))
   ....:     

In [63]: !cat result.csv
100000,15,foo
200000,25,bar
300000,35,baz