CODEWITHSUNDEEP
python
elitecheese1337
elitecheese1337

Reputation: 179

How to pick out specific parts of a list?

So I have this list:

Comps = [
    [1000, 500, 200, 100, 700, 350, 120, 900, 800],
    [1001, 501, 201, 101, 701, 351, 121, 901, 801]
    ]

How would I get the results [500, 350, 120] and [501, 351, 121] from this list?

I know how to get the answer [500, 350, 120] if there was only the first list (which would be [1000, 500, 200, 100, 700, 350, 120, 900, 800]) in the variable Comps. However, I am not sure how to work with a list that contains 2 or more lists (also known as "list of list"). I keep getting the error: "IndexError: list index out of range".

What I've done is this:

print(list(Comps[i] for i in [1, 5, 6]))

Here is the full error for those who are interested:

    Traceback (most recent call last):
  File "<pyshell#74>", line 1, in <module>
    print(list(Comps[i] for i in [1, 5, 6]))
  File "<pyshell#74>", line 1, in <genexpr>
    print(list(Comps[i] for i in [1, 5, 6]))
IndexError: list index out of range

Upvotes: 1

Views: 157

Answers (3)

Eric Duminil
Eric Duminil

Reputation: 54223

The reason you get a IndexError: list index out of range is that Comp is a list with only 2 elements. Element at index 1 is the second list ([1001, 501, 201, 101, 701, 351, 121, 901, 801]) but there's no element at index 5 in Comps.

You could use a nested list comprehension:

>>> Comps = [[1000, 500, 200, 100, 700, 350, 120, 900, 800], [1001, 501, 201, 101, 701, 351, 121, 901, 801]]
>>> [[l[i] for i in [1, 5, 6]] for l in Comps]
[[500, 350, 120], [501, 351, 121]]

If you want them in a flat list:

>>> [l[i] for l in Comps for i in [1, 5, 6]]
[500, 350, 120, 501, 351, 121]

If you work with matrices and arrays, you might want to take a look at numpy:

import numpy as np
comps = np.array([[1000, 500, 200, 100, 700, 350, 120, 900, 800], [1001, 501, 201, 101, 701, 351, 121, 901, 801]])
comps[:,[1, 5, 6]]
# array([[500, 350, 120],
#       [501, 351, 121]])

If you only want one of them:

>>> [Comps[0][i] for i in [1, 5, 6]]
[500, 350, 120]
>>> [Comps[1][i] for i in [1, 5, 6]]
[501, 351, 121]

And with NumPy:

>>> comps[0, [1, 5, 6]]
array([500, 350, 120])
>>> comps[1, [1, 5, 6]]
array([501, 351, 121])

Upvotes: 3

simkusr
simkusr

Reputation: 812

As there is one list like:

1) math=[one, list] you always use mathList = math[theIndexOfItemInList]

but as there is two or more you need to change how you count lists, like if there is a list:

2) math=[[one, list], [second, list], [third, list], [etc, list]]

as you call this list:

3) mathList = math[whichListYouNeed][theIndexOfItemInList]

Upvotes: 1

juanpa.arrivillaga
juanpa.arrivillaga

Reputation: 95948

You are almost there, you just need to nest one level deeper:

[[sub[i] for i in [1, 5, 6]] for sub in Comps]

Alternatively, you can use operator.itemgetter:

>>> list(map(operator.itemgetter(*[1, 5, 6]), Comps))
[(500, 350, 120), (501, 351, 121)]

Or, more directly, 

>>> list(map(operator.itemgetter(1, 5, 6), Comps))
[(500, 350, 120), (501, 351, 121)]

Upvotes: 1

Related Questions