Stackable Position Adjustments

import pandas as pd

from lets_plot import *

LetsPlot.setup_html()

df = pd.read_csv("https://raw.githubusercontent.com/JetBrains/lets-plot-docs/master/data/mpg.csv")
print(df.shape)
df.head()

(234, 12)

	Unnamed: 0	manufacturer	model	displ	year	cyl	trans	drv	cty	hwy	fl	class
0	1	audi	a4	1.8	1999	4	auto(l5)	f	18	29	p	compact
1	2	audi	a4	1.8	1999	4	manual(m5)	f	21	29	p	compact
2	3	audi	a4	2.0	2008	4	manual(m6)	f	20	31	p	compact
3	4	audi	a4	2.0	2008	4	auto(av)	f	21	30	p	compact
4	5	audi	a4	2.8	1999	6	auto(l5)	f	16	26	p	compact

1. Default Presentation of 'stack' and 'fill'

Stacking is the default behaviour for most area plots. Fill makes it easier to compare proportions.

ggplot(df, aes(x="drv", fill="year", group="year")) + \
    geom_bar() + \
    scale_fill_discrete(format="d") + \
    ggtitle("bar: position='stack' (default)")

ggplot(df, aes(x="drv", fill="year", group="year")) + \
    geom_bar(position='fill') + \
    scale_fill_discrete(format="d") + \
    ggtitle("bar: position='fill'")

ggplot(df, aes(x="hwy", fill="year", group="year")) + \
    geom_histogram() + \
    scale_fill_discrete(format="d") + \
    ggtitle("histogram: position='stack' (default)")

ggplot(df, aes(x="hwy", fill="year", group="year")) + \
    geom_histogram(position='fill') + \
    scale_fill_discrete(format="d") + \
    ggtitle("histogram: position='fill'")

ggplot(df, aes(x="hwy", fill="year", group="year")) + \
    geom_area(stat='density', color="black") + \
    scale_fill_discrete(format="d") + \
    ggtitle("area: position='stack' (default)")

ggplot(df, aes(x="hwy", fill="year", group="year")) + \
    geom_area(stat='density', position='fill', color="black") + \
    scale_fill_discrete(format="d") + \
    ggtitle("area: position='fill'")

2. Stacking Order

Control the stacking order by changing the order parameter of the as_discrete() function.

ggplot(df, aes(x="hwy")) + \
    geom_area(aes(fill=as_discrete("drv")), stat='density', color="black") + \
    ggtitle("Default order of drive types")

ggplot(df, aes(x="hwy")) + \
    geom_area(aes(fill=as_discrete("drv", order=-1)), stat='density', color="black") + \
    ggtitle("Backward order of drive types")

3. Other Geometries

Let’s have a look at the geometries, for which the default position adjustment differs from 'stack' and 'fill'.

When stacking across multiple layers it’s a good idea to set the group aesthetic - this ensures that all layers are stacked in the same way.

ggplot(df, aes(x="drv", color="year", group="year")) + \
    geom_line(stat='count', position='stack') + \
    geom_point(stat='count', position='stack') + \
    scale_color_discrete(format="d") + \
    ggtitle("Line and point")

ggplot(df, aes(x="drv")) + \
    geom_area(aes(fill="year", group="year"), stat='count', position='stack', color="black") + \
    geom_line(aes(group="year"), stat='count', position='stack', color="black") + \
    scale_fill_discrete(format="d") + \
    ggtitle("Area and line")

4. Parameter vjust

The vjust argument of position_stack() is used to move the location of plot elements vertically (vjust stands for vertical adjustment).

ggplot(df, aes(x="drv")) + \
    geom_area(aes(fill="year", group="year"), stat='count', \
              position='stack', color="black") + \
    geom_label(aes(label="..count..", group="year"), stat='count', \
               position='stack', color="black") + \
    scale_fill_discrete(format="d") + \
    ggtitle("vjust=1 (default)")

ggplot(df, aes(x="drv")) + \
    geom_area(aes(fill="year", group="year"), stat='count', \
              position='stack', color="black") + \
    geom_label(aes(label="..count..", group="year"), stat='count', \
               position=position_stack(vjust=.5), color="black") + \
    scale_fill_discrete(format="d") + \
    ggtitle("vjust=0.5")

5. Parameter mode

By default only objects from different groups are stacked over each other, and objects inside one group are positioned as in position='identity'.

This behaviour could be changed by switching mode parameter.

mode='all' means, that each object will be shifted.

ggplot(df, aes(x="hwy", color="class")) + \
    geom_point(y=1, position=position_stack()) + \
    coord_fixed() + ylim(1, 35) + \
    ggtitle("mode='groups' (default)")

ggplot(df, aes(x="hwy", color="class")) + \
    geom_point(y=1, position=position_stack(mode='all')) + \
    coord_fixed() + ylim(1, 35) + \
    ggtitle("mode='all'")

6. Negative Values

Stacking supports positive and negative shifts of the objects separately.

data = {
    "x": [0, 1, 2, 0, 1, 2],
    "y": [4, 2, 3, 0, 1, -1],
    "g": ["a", "a", "a", "b", "b", "b"],
}

ggplot(data, aes("x", "y", fill="g")) + \
    geom_bar(stat='identity') + \
    geom_hline(yintercept=0, color="black")

ggplot(data, aes("x", "y")) + \
    geom_bar(aes(fill="g"), stat='identity', position='fill') + \
    geom_label(aes(label="g", group="g"), stat='identity', \
               position=position_fill(vjust=.5)) + \
    geom_hline(yintercept=0, color="black")