current community

your communities

more stack exchange communities

Stack Exchange
tour help
stack overflow careers
Take the 2-minute tour ×
Stack Overflow is a question and answer site for professional and enthusiast programmers. It's 100% free, no registration required.
up vote 1 down vote favorite

I am trying to plot a boxplot for a column in several csv files (without the header row of course), but running into some confusion around tuples, lists and arrays. Here is what I have so far

    #!/usr/bin/env python

    import csv
    from numpy import *
    import pylab as p
    import matplotlib

    #open one file, until boxplot-ing works
    f = csv.reader (open('2-node.csv'))
    #get all the columns in the file
timeStamp,elapsed,label,responseCode,responseMessage,threadName,dataType,success,bytes,Latency = zip(*f)

    #Make list out of elapsed to pop the 1st element -- the header
    elapsed_list = list(elapsed)
    elapsed_list.pop(0)

    #Turn list back to a tuple
    elapsed = tuple(elapsed_list)

    #Turn list to an numpy array 
    elapsed_array = array(elapsed_list)

    #Elapsed Column statically entered into an array
    data_array = ([4631, 3641, 1902, 1937, 1745, 8937] )

    print data_array #prints in this format: ([xx,xx,xx,xx]), .__class__ is list ... ?
    print elapsed    #prints in this format: ('xx','xx','xx','xx'), .__class__ is tuple
    print elapsed_list # #print in this format: ['xx', 'xx', 'xx', 'xx', 'xx'], .__class__ is list
    print elapsed_array #prints in this format: ['xx' 'xx' 'xx' 'xx' 'xx'] -- notice no commas, .__class__ is numpy.ndarray

    p.boxplot (data_array) #works
    p.boxplot (elapsed) # does not work, error below
    p.boxplit (elapsed_list) #does not work
    p.boxplot (elapsed_array) #does not work
    p.show()

For boxplots, the 1st argument is an "an array or a sequence of vectors", so I would think elapsed_array would work ... ? But yet data_array, a "list," works ... but elapsed_list` a "list" does not ... ? Is there a better way to do this ... ?

I am fairly new to python, and I would like to understand the what about the differences among a tuple, list, and numpy-array prevents this boxplot from working.

Example error message is:

Traceback (most recent call last):
  File "../pullcol.py", line 32, in <module>
    p.boxplot (elapsed_list)
  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/pyplot.py", line 1962, in boxplot
    ret = ax.boxplot(x, notch, sym, vert, whis, positions, widths, patch_artist, bootstrap)
  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/axes.py", line 5383, in boxplot
    q1, med, q3 = mlab.prctile(d,[25,50,75])
  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/mlab.py", line 946, in prctile
    return _interpolate(values[ai],values[bi],frac)
  File "/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/matplotlib/mlab.py", line 920, in _interpolate
    return a + (b - a)*fraction
TypeError: unsupported operand type(s) for -: 'numpy.ndarray' and 'numpy.ndarray'
share|improve this question

2 Answers 2

up vote 4 down vote accepted

elapsed contains strings. Matplotlib needs integers or floats to plot something. Try converting each value of elapsed to integer. You can do this like so 

elapsed = tuple([int(i) for i in elapsed])

or as FredL commented below:

elapsed_list = array(elapsed_list, dtype=float)
share|improve this answer
2  
Was about to say the same thing - basically create the array using something like "array(elapsed_list, dtype=float)" –  FredL Aug 31 '11 at 14:53
    
That did it! So how do I tell if elapsed contains strings or something else ... ? Is it the 'xx' which is a string ... ? They all look like numbers (-: –  KM. Aug 31 '11 at 14:59
2  
@KM, if you look at the type of one of the elements of elapsed it will tell you if it's a string: print type(elapsed[0]) prints <type 'str'>; if you add the line print type(data_array[0]), you will see instead <type 'int'>. And yes, the 'xx' means it's a string, but the type method is explicit. Does this answer your question? –  Yann Aug 31 '11 at 15:03
    
Thanks @Yann, it does, thank you once again! –  KM. Aug 31 '11 at 15:10
up vote 1 down vote

I'm not familiar with numpy or matplotlib, but just from the description and what's working, it appears it is looking for a nested sequence of sequences. Which is why data_array works as it's a tuple containing a list, where as all your other input is only one layer deep.

As for the differences, a list is a mutable sequence of objects, a tuple is an immutable sequence of objects and an array is a mutable sequence of bytes, ints, chars (basically 1, 2, 4 or 8 byte values).

Here's a link to the Python docs about 5.6. Sequence Types, from there you can jump to more detailed info about lists, tuples, arrays or any of the other sequence types in Python.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.