Applying color gradients to columns

Numeric trends in larger tables can quickly become difficult to see. Adding color to reveal these trends can be one way to make reading complex tables a little easier.

The data

Clean gradients in real data can be hard to come by. Excuse the contrived nature of the data below, but I really want to make a rainbow gradient!

using StyledTables, DataFrames, Colors

df = DataFrame(id = 1:8, score = rand(8))

8×2 DataFrame

Row	id	score
	Int64	Float64
1	1	0.785614
2	2	0.852325
3	3	0.906774
4	4	0.0540956
5	5	0.432241
6	6	0.0484888
7	7	0.596142
8	8	0.620939

So far so boring.

The gradient we'll be applying is the following:

colors = range(HSV(0,1,1), stop = HSV(-360,1,1), length = nrow(df))

Step 1: Styling function

The function below is more general than what would be required for the specific example here. In our case, it would be possible to simply use the id variable to index into colors.

function apply_gradient(x, xmax; colors)
    color = colors[round(Int, (x / xmax) * length(colors))]
    return (; color)
end

apply_gradient (generic function with 1 method)

Step 2: Apply styling

tbl = StyledTable(df)
tab_style!(x -> apply_gradient(x, maximum(df.id); colors), tbl, :id)
render(tbl)

id	score
1	0.786
2	0.852
3	0.907
4	0.0541
5	0.432
6	0.0485
7	0.596
8	0.621

Step 3: Column labels

cols_label!(tbl, :id => "Student ID", :score => "Score")
render(tbl)

Student ID	Score
1	0.786
2	0.852
3	0.907
4	0.0541
5	0.432
6	0.0485
7	0.596
8	0.621