Manipulating Date Breaks and Labels

from plotnine import (
    ggplot,
    aes,
    geom_point,
    scale_x_datetime,
    labs,
    theme_set,
    theme_linedraw,
)
from plotnine.data import economics

from mizani.breaks import breaks_date_width
from mizani.formatters import label_date

theme_set(theme_linedraw())  # default theme

Using the economics dataset

economics.head()

	date	pce	pop	psavert	uempmed	unemploy
0	1967-07-01	507.4	198712	12.5	4.5	2944
1	1967-08-01	510.5	198911	12.5	4.7	2945
2	1967-09-01	516.3	199113	11.7	4.6	2958
3	1967-10-01	512.9	199311	12.5	4.9	3143
4	1967-11-01	518.1	199498	12.5	4.7	3066

How does the saving rate vary with time?

(
    ggplot(economics)
    + geom_point(aes("date", "psavert"))
    + labs(y="personal saving rate")
)

Yikes! the calculated breaks are awful, we need to intervene. We do so using the breaks_date_width and label_date

functions from mizani.

Set breaks every 10 years

(
    ggplot(economics)
    + geom_point(aes("date", "psavert"))
    + scale_x_datetime(breaks=breaks_date_width("10 years"))  # new
    + labs(y="personal saving rate")
)

That is better. Since all the breaks are at the beginning of the year, we can omit the month and day. Using label_date we override the format string. For more on the options for the format string see the strftime behavior.

(
    ggplot(economics)
    + geom_point(aes("date", "psavert"))
    + scale_x_datetime(
        breaks=breaks_date_width("10 years"), labels=label_date("%Y")
    )  # modified
    + labs(y="personal saving rate")
)

We can achieve the same result with a custom formating function.

def custom_label_date1(breaks):
    """
    Function to format the date
    """
    return [x.year if x.month == 1 and x.day == 1 else "" for x in breaks]


(
    ggplot(economics)
    + geom_point(aes("date", "psavert"))
    + scale_x_datetime(  # modified
        breaks=breaks_date_width("10 years"), labels=custom_label_date1
    )
    + labs(y="personal saving rate")
)

We can use a custom formatting function to get results that are not obtainable with the label_date function. For example if we have monthly breaks over a handful of years we can mix date formats as follows; specify beginning of the year and every other month. Such tricks can be used reduce overcrowding.

from datetime import date


def custom_label_date2(breaks):
    """
    Function to format the date
    """
    res = []
    for x in breaks:
        # First day of the year
        if x.month == 1 and x.day == 1:
            fmt = "%Y"
        # Every other month
        elif x.month % 2 != 0:
            fmt = "%b"
        else:
            fmt = ""

        res.append(date.strftime(x, fmt))

    return res


(
    ggplot(economics.loc[40:60, :])  # modified
    + geom_point(aes("date", "psavert"))
    + scale_x_datetime(
        breaks=breaks_date_width("1 months"), labels=custom_label_date2, minor_breaks=[]
    )
    + labs(y="personal saving rate")
)

We removed the labels but not the breaks, leaving behind dangling ticks for the skipped months. We can fix that by wrapping breaks_date_width around a filtering function.

def custom_label_date3(breaks):
    """
    Function to format the date
    """
    res = []
    for x in breaks:
        # First day of the year
        if x.month == 1:
            fmt = "%Y"
        else:
            fmt = "%b"

        res.append(date.strftime(x, fmt))

    return res


def custom_breaks_date_width(width=None):
    """
    Create a function that calculates date breaks

    It delegates the work to `breaks_date_width`
    """

    def filter_func(limits):
        breaks = breaks_date_width(width)(limits)
        # filter
        return [x for x in breaks if x.month % 2]

    return filter_func


(
    ggplot(economics.loc[40:60, :])
    + geom_point(aes("date", "psavert"))
    + scale_x_datetime(  # modified
        breaks=custom_breaks_date_width("1 months"), labels=custom_label_date3
    )
    + labs(y="personal saving rate")
)

The breaks and labels functions are tightly coupled to give us exactly what we want.

Credit: This example was motivated by the github user lorin (Lorin Hochstein) and his endeavor to control date breaks and date labels.