numpy beginner: writing an array using numpy.savetxt

Question

I have a numpy histogram that I would like to output as a tab-delimited text file. My code is below:

targethist = np.histogram(targetlist, bins=ilist)
print targethist
np.savetxt('ChrI_dens.txt',targethist,delimiter='\t')

targetlist and ilist are long lists of integers. I get the following output:

(array([0, 0, 0, ..., 0, 0, 0]), array([ 1, 10000, 20000, ..., 15060000, 15070000, 15072422])) Traceback (most recent call last): File "target_dens_np.py", line 62, in np.savetxt('ChrI_dens.txt',targethist,delimiter='\t') File "/Library/Frameworks/Python.framework/Versions/7.3/lib/python2.7/site-packages/numpy/lib/npyio.py", line 979, in savetxt fh.write(asbytes(format % tuple(row) + newline)) TypeError: float argument required, not numpy.ndarray

It seems that the histogram array has been created, but I have done something wrong in the np.savetxt() line. I have read the documentation, but don't understand why any of the arguments in this function would be expecting a float. Where have I gone wrong?

mgilson · Accepted Answer · 2012-06-06 19:06:02Z

5

I think that the problem is that the second argument to savetxt must be "array-like". Your input is not "array-like". e.g.

print (len(targethist[0]))
print (len(targethist[1]))

Notice the lengths aren't the same? If the lengths were the same, numpy could convert it to a single 2-D array and everything would be fine, but it can't do the conversion so it fails.

This works

np.savetxt('stuff.dat',(targethist[0],targethist[1][1:]),delimiter='\t')

But I've truncated your data ;). You'll need to decide what you want to do to work around this one.

I must admit, the error message here is quite cryptic.

answered Jun 6, 2012 at 19:06

mgilson

312k70 gold badges656 silver badges722 bronze badges

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

2 Comments

pandaSeq Over a year ago

Spot on! This should have occurred to me before I encountered the cryptic error, anyway. As it turns out, I want to truncate the first integer in the longer array, so this is perfect. Thanks!

mgilson Over a year ago

@pandaSeq I have to admit -- based on that traceback I was scratching my head on this one for a few minutes. Usually when python-numpy questions sit on SO for more than 10 minutes without any answers, you know something funny is going on ;).

Collectives™ on Stack Overflow

numpy beginner: writing an array using numpy.savetxt

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related