How to select a column name and use it as input for a variable name in Python?

Question

ORIGINAL QUESTION:

I'm writing a while loop to loop over certain columns. In this while loop I want to create a variable of which the name partly consists of the column name it is looping over.

x=2
length=len(grouped_class.columns)
while x<length:
    x=x+1
    (grouped_class.columns[x])_largest = x+5
    ...

This is my current code (=x+5 is not actual code, but as example), but it returns a syntax error. If I run grouped_class.columns[x] in the shell it returns the name of that column, for example : "ColumnA". I want to use this "ColumnA" as first part of a variable name.

So in the variable list it would return: ColumnA_largest

In this way I can store the result for each column in a seperate variable.

How can I do this?

EDIT: QUESTION GENERALIZED

How can I use a string obtained by df.column[x] as input for a variable name?

Example df:

ColumnA    ColumnB    ColumnC
5          6          4
6          10         2

If I run df.columns[1] it returns "ColumnB"

I want to use this "ColumnB" as part of the name when assigning a variable.

Imagine I want to create the variable COLNAME_sum = x + 5 I would like to change the COLNAME to the string I obtained from df.columns[1] (="ColumnB")

Expected output: A variable named ColumnB_sum.

How can I do this?

Can you show example of input and expected output? If you need help look at how to provide minimal reproducible example. — zipa
– zipa, Commented Jun 26, 2019 at 11:43
It is a complicated dataset and I did not eleborate on that because it would possible be distracting. I also think it is not relevant for the question (perhaps I'm wrong). I would just like to know how to use a string value as part of a variable name.I can edit the question to make it more general. — RobH
– RobH, Commented Jun 26, 2019 at 11:47
@Robvh my answer below explains how it can be done explicitly, but should be avoided. — WiseDev
– WiseDev, Commented Jun 26, 2019 at 11:57

zipa · Accepted Answer · 2019-06-26 12:49:43Z

1

You shouldn't create variables on-the-fly as it could lead to many issues, instead, use dictionary:

largest = {}
x = 2
length = len(grouped_class.columns)
while x < length:
    x = x + 1
    column = grouped_class.columns[x]
    largest[column + '_largest'] = x + 5
    ...

edited Jun 26, 2019 at 12:49

answered Jun 26, 2019 at 11:51

zipa

28k6 gold badges45 silver badges62 bronze badges

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

3 Comments

RobH Over a year ago

Thanks, this basiscally does what I want, but I'm stumbeling upon a new problem. In my real code I've now written:

largest={} x=2 length=len(grouped_class.columns)-1 while x<length:     x=x+1     largest[grouped_class.columns[x]+"_largest"]=grouped_class.nlargest(5,[grouped_class.columns[x]])

. So for each grouped class column I want the rows with 5 largest values. However, it now makes a dictionary with the correct keys COLNAME_largest but the values are wrong. It returns 5 values for every column and not just the selected one. Only the values of the selected one are sorted.

RobH Over a year ago

for each key (= column name) I want only like the sorted values of that column name.

zipa Over a year ago

@Robvh I've added part with column variable so you can use grouped_class[column].nlargest(5) or even add .tolist() at the end of it.

WiseDev · Accepted Answer · 2019-06-26 11:53:43Z

1

Right way: The right (Pythonic) way is to use dictionaries.

columns = {}
columns[some_string] = some_value

Unadvised dirty way, but answers your question: Storing a string as a variable name in your global namespace can be done simply by (example):

some_value = 100
some_string = 'var_name'
globals()[some_string] = some_value

The output is then

>>> var_name
100

On the other hand, if you want to add a variablename locally, you can use locals() instead of globals().

I trust you can take over from here!

edited Jun 26, 2019 at 11:53

answered Jun 26, 2019 at 11:43

WiseDev

5854 silver badges19 bronze badges

1 Comment

WiseDev Over a year ago

I edited my answer to make it more compliant to general expectations.

Nordle · Accepted Answer · 2019-06-26 11:58:34Z

1

You do not want to do that. Of course dirty tricks can allow it, but the Pythonic way is to use a dictionary:

largest = {}
x = 2
length = len(grouped_class.columns)
while x < length:
    x = x + 1
    largest[grouped_class.columns[x]] = x + 5

edited Jun 26, 2019 at 11:58

Nordle

3,0013 gold badges18 silver badges37 bronze badges

answered Jun 26, 2019 at 11:52

Serge Ballesta

150k13 gold badges137 silver badges267 bronze badges

Comments

amalik2205 · Accepted Answer · 2019-06-26 12:12:59Z

1

Looks like you are using a pandas dataFrame. You can use:

dict = {}
my_dict[grouped_class.columns[x]+'_largest'] = x+5

answered Jun 26, 2019 at 12:12

amalik2205

4,2031 gold badge17 silver badges21 bronze badges

Collectives™ on Stack Overflow

How to select a column name and use it as input for a variable name in Python?

4 Answers 4

3 Comments

1 Comment

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

3 Comments

1 Comment

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related