Pandas Styling - Coloring cells of specific columns instead of entire DataFrame

Question

I wrote the following block of code to colour some cells in my dataframe.

def applycolor(dtf):

        return ['background-color: lightgreen' if x >= 1500  else 
        ('background-color: lightskyblue' if 1000 <= x < 1500  else 
        'background-color: pink' if 750 <= x < 1000 else
        'background-color: wheat' if  550 <= x < 750 else 
        'background-color: paleturquoise' if 330 <= x < 550  else 
        'background-color: darkseagreen' if 150 <= x < 330 else 'background-color: default') for x in dtf]    


cs1 = cs.style.apply(applycolor, axis=0)

This gave me a result as shown in the image.

However, I only want the colors to be rendered for the figures specified in df['$-Score']. But this styling appended colours to all the relevant numerics of the dataframe as seen.

I tried to change last line of the list comprehension to include only the particular column of the dataframe like so: .....if 150 <= x < 330 else 'background-color: default') for x in dtf['$-Score']- but it returned an error.

Have tried looking for the specific answer, but haven't been able to find it. Any ideas?

Alternatively, a SAMPLE DATAFRAME:

    A    B   C
0   83  76  30
1   34  17  44
2   72  16  94
3   24  94  27
4   98  36  35
5   41  77  39
6   65  54  18
7   85  1   59
8   12  79  2
9   33  57  76
10  66  69  100
11  99  51  89
12  24  74  32
13  51  98  63
14  63  36  82
15  53  52  65

I only want numbers between 55 and 58 to be coloured red in column B and Blue between 84 and 87 in column C only.

How can I go about it?

Won't be possible to provide the same dataframe honestly. But you can make a random one with 3 columns A B C with numerical data ( random numbers in range 0 to 100) in each and highlight all numbers between 50 and 55 in column B only and between 84 and 87 in column C only. You'll get a better idea that way. — therion
– therion, Commented Feb 11, 2020 at 18:16
exactly what you said, can you add that in the question as a sample? on which your function would work — anky
– anky, Commented Feb 11, 2020 at 18:18

wkzhu · Accepted Answer · 2020-02-11 22:07:37Z

2

Style.apply works just like DataFrame.apply in that the dataframe is broken into series, and in your case you want to do something different based on each series's name (i.e., column name). So the below example can be extended for your purposes:

def apply_formatting(col):
    if col.name == 'a':
        return ['background-color: red' if c > 50 else '' for c in col.values]
    if col.name == 'b':
        return ['background-color: green' if c > 10 else '' for c in col.values]
    if col.name == 'c':
        return ['background-color: blue' if c > 30 else '' for c in col.values]

data = pd.DataFrame(
    np.random.randint(0, 100, 30).reshape(10, 3), columns=['a', 'b', 'c'])

data.style.apply(apply_formatting)  # axis=0 by default

Here is the result on the random data:

answered Feb 11, 2020 at 22:07

wkzhu

1,65015 silver badges24 bronze badges

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

2 Comments

therion Over a year ago

This works for the sample. But when I add .name == 'xyz' I get a reshape error. Why is that happening?

therion Over a year ago

I found the solution. The real solution was using 'subset' = {column name} in the df.style syntax.

Oleks · Accepted Answer · 2021-03-02 13:44:08Z

1

Just use subset property


df.style.apply(your_func, subset=['column_name'])

answered Mar 2, 2021 at 13:44

Oleks

314 bronze badges

Collectives™ on Stack Overflow

Pandas Styling - Coloring cells of specific columns instead of entire DataFrame

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related