Pandas pct_change(): Add Colors to Percent with style

Add Annotations to Pandas pct_change results
Add Annotations to Pandas pct_change results

Pandas pct_change() function is a handy function that lets us calculate percent change between two rows or two columns easily. By default, pct_change() function works with adjacent rows and columns, but it can compute percent change for user defined period as well.

One of the useful features of Pandas pct_change is to add annotation with multiple colors using Pandas style method. In this tutorial, we will learn how to add colors to results obtained from using Pandas pct_change() function to compute percent change between rows.

Add Annotations to Pandas pct_change results

First, let us load Pandas.

import pandas as pd

We will create a simple dataframe using three tech companies’ revenue over multiple years.

year=[2017, 2018, 2019, 2020]
facebook =[15934, 22112, 18485, 29146]
google= [12662, 30736, None, 40269]
microsoft= [25489, 16571, 39240, 44281]

Our input data is stored in multiple lists and we can convert the lists into a dataframe using Pandas’ DataFrame() function.

df = pd.DataFrame({"facebook":facebook,
                   "google": google,
                   "microsoft": microsoft},
                 index=year)

In the example dataframe columns are companies and rows are the years.

df
facebook	google	microsoft
2017	15934	12662.0	25489
2018	22112	30736.0	16571
2019	18485	NaN	39240
2020	29146	40269.0	44281

If we want to compute the change in revenue over time in terms of percent, We can use Pandas’ pct_change() function on the dataframe. By default, Pandas pct-change() function computes percent change for every row by comparing it with the previous row. That is why in the results, the first row values are NaNs.

In our example, we get percent change in revenue for every year.

df.pct_change()


facebook	google	microsoft
2017	NaN	NaN	NaN
2018	0.387724	1.427421	-0.349876
2019	-0.164029	0.000000	1.367992
2020	0.576738	0.310157	0.128466

Add Percentage Sign in Pandas

We can add percentage symbol to the results from pct_change() using style method and specify the format that we would like to have.

df.pct_change().style.format("{:.2%}")

facebook	google	microsoft
2017	nan%	nan%	nan%
2018	38.77%	142.74%	-34.99%
2019	-16.40%	0.00%	136.80%
2020	57.67%	31.02%	12.85%

Note that the row with nan values also have percentage sign and that does not make sense. We can change the values of nan using “na_rep” to format() function’s argument. Now we get “-” dashes instead of nan with percentage symbol.

df.pct_change().style.format("{:.2%}", na_rep="-")

	facebook	google	microsoft
2017	-	-	-
2018	38.77%	142.74%	-34.99%
2019	-16.40%	0.00%	136.80%
2020	57.67%	31.02%	12.85%

Annotate Maximum Values in a Column with colors in Pandas

To highlight maximumum values We can highlight maximum values in each column, we can use highlight_max() function after converting to percent change using using chain operator.

(df.
 pct_change().
 style.
 highlight_max().
 format("{:.2%}", na_rep="-"))

Note the difference in the way we chained multiple functions. When you combining multiple operations, writing each operation in a separate line as here makes it easy to read the code and understand.
By default, highlight_max() function annotates the maximum values in each column in yellow color.

Highlight Max Value with highlight_max

We can also specify the color with which we would like to highlight maximum value using color argument to highlight_max() function.

(df.
 pct_change().
 style.
 highlight_max(color="lightgreen").
 format("{:.2%}", na_rep="-"))
Specify Color to highlight Max Value with highlight_max

Annotate Maximum and Minimum Values in a Column with colors in Pandas

Let us highlight both maximum value and minimum value in each column with two different colors using highlight_max/highlight_min functions.

(df.
 pct_change().
 style.
 highlight_max(color="lightgreen").
 highlight_min(color="yellow").
 format("{:.2%}", na_rep="-"))
Pandas add colors to maximum and minimum values