Geek Answers Handbook

For a loop for second items only in a list of lists. (Python)

What i have

something like that

def mymethod():
    return [[1,2,3,4],
            [1,2,3,4],
            [1,2,3,4],
            [1,2,3,4]]

mylist = mymethod()

for _, thing, _, _ in mylist:
    print thing

# this bit is meant to be outside the for loop, 
# I mean it to represent the last value thing was in the for
if thing:
    print thing

What I want

what i want to do is to avoid dummy variables, is there a smarter way to do this than

for thing in mylist:
    print thing[1]

because then I would have to use thing[1]any other time I needed without assigning it a new variable, and then everything becomes messy.

newish for python so sorry if i miss something obvious

+5
source share
5 answers

You can crack the generator expression

def mymethod():
    return [[1,2,3,4],
            [1,2,3,4],
            [1,2,3,4],
            [1,2,3,4]]

mylist = mymethod()

for thing in (i[1] for i in mylist):
    print thing

# this bit is meant to be outside the for loop, 
# I mean it to represent the last value thing was in the for
if thing:
    print thing
+7
source

If you want to get the second column of an array, you can use list comprehension, for example:

a = [ [ 1, 2, 3, 4 ],
      [ 5, 6, 7, 8 ],
      [ 9,10,11,12 ],
      [13,14,15,16 ] ]


second_column = [ row[1] for row in a ]
# you get [2, 6, 10, 14]

You can wrap this with a function:

def get_column ( array, column_number ):
    try:
        return [row[column_number] for row in array]
    except IndexError:
        print ("Not enough columns!")
        raise # Raise the exception again as we haven't dealt with the issue.

fourth_column = get_column(a,3)
# you get [4, 8, 12, 16]

tenth_column = get_column(a,9)
# You requested the tenth column of a 4-column array, so you get the "not enough columns!" message.

, , numpy , .

, Lattyware , :

def column_iterator ( array, column_number ):
    try:
        for row in array:
            yield row[column_number]
    except IndexError:
        print ("Not enough columns!")
        raise

:

>>> for item in column_iterator(a,1):
...    print(item)
... 
2
6
10
14
>>>

:

>>> b = column_iterator(a,1)
>>> b.next()
2
>>> b.next()
6
>>> b.next()
10
>>> b.next()
14
+3

, itertools.chain slicing ?

for thing in itertools.islice(itertools.chain(*mylist),1,None,len(mylist)):
    print(thing)

Numpy . numpy

for thing in numpy.array(mylist)[:,1]:
    print(thing)
+2

Dikei , , :

for sublist in mylist:
    item = sublist[1]
    ...
    do_stuff(item)
    ...
    do_other_stuff(item)
    ...

, , , , , .

- , , , , , :

python -m timeit -s "mylist = [range(1,8) for _ in range(1,8)]" 'for thing in mylist:' '    item=thing[1]' '    pass'
1000000 loops, best of 3: 1.25 usec per loop

python -m timeit -s "mylist = [range(1,8) for _ in range(1,8)]" 'for thing in (i[1] for i in mylist):' '    pass'
100000 loops, best of 3: 2.37 usec per loop

python -m timeit -s "mylist = [range(1,8) for _ in range(1,8)]" 'for thing in itertools.islice(itertools.chain(*mylist),1,None,len(mylist)):' '    pass'
1000000 loops, best of 3: 2.21 usec per loop

python -m timeit -s "import numpy" -s "mylist = numpy.array([range(1,8) for _ in range(1,8)])" 'for thing in mylist[:,1]:' '    pass' 
1000000 loops, best of 3: 1.7 usec per loop

python -m timeit -s "import numpy" -s "mylist = [range(1,8) for _ in range(1,8)]" 'for thing in numpy.array(mylist)[:,1]:' '    pass'
10000 loops, best of 3: 63.8 usec per loop

, numpy , , , .

:

python -m timeit -s "mylist = [range(1,100) for _ in range(1,100)]" 'for thing in mylist:' '    item=thing[1]' '    pass'
100000 loops, best of 3: 16.3 usec per loop

python -m timeit -s "mylist = [range(1,100) for _ in range(1,100)]" 'for thing in (i[1] for i in mylist):' '    pass'
10000 loops, best of 3: 27 usec per loop

python -m timeit -s "mylist = [range(1,100) for _ in range(1,100)]" 'for thing in itertools.islice(itertools.chain(*mylist),1,None,len(mylist)):' '    pass'
10000 loops, best of 3: 101 usec per loop

python -m timeit -s "import numpy" -s "mylist = numpy.array([range(1,100) for _ in range(1,100)])" 'for thing in mylist[:,1]:' '    pass'
100000 loops, best of 3: 8.47 usec per loop

python -m timeit -s "import numpy" -s "mylist = [range(1,100) for _ in range(1,100)]" 'for thing in numpy.array(mylist)[:,1]:' '    pass'
100 loops, best of 3: 3.82 msec per loop

Remember that speed should always be second for readability if you do not need it.

+1
source

To solve this problem, you can use the method itemgetter():

from operator import itemgetter

def mymethod():
    return [[1,2,3,4],
            [1,2,3,4],
            [1,2,3,4],
            [1,2,3,4]]

mylist = mymethod()

row = map(itemgetter(2), mylist)
print("row %s" % row)

thing = row[-1]

# this bit is meant to be outside the for loop, 
# I mean it to represent the last value thing was in the for
if thing:
    print thing

Conclusion:

row [3, 3, 3, 3]
3
+1
source

More articles:


All Articles