Skip to content

A gallery of static Python plots that the Big Data Innovation Team has produced

License

Notifications You must be signed in to change notification settings

CityofToronto/bdit_plotting_gallery

Repository files navigation

Welcome to the BDITTO Plotting Gallery !

Are you tired of re-inventing the matplotlib wheel every time you need a chart for a new project? Did you forget where to find the right colour palette for those bars? Do you need a total rewrite just to include one more line in a line chart? If you said yes to any of the above, this Gallery is for you!

This repo contains a collection of reusable chart functions that strive to conform to standardized formats and designs based on output the BDITTO team has already created for various projects.

The Gallery collection includes:

  • a built-in module called Repeatable Information Charts Kit (RICK) that already contains many chart types for use out-of-the-box. Some customization is possible. The source code for all RICK charts is found in rick.py.
  • stand-alone examples beyond the scope of RICK with source code up-front to allow easy customization as you desire

All charts were developed with proper python documentation to allow the documentation for each chart to be auto-generated using sphinx.

Contribute to the Gallery

This is a work in progress and we need YOU to add to the collective knowledge! Please go to the README in the sphinx directory for a step-by-step guide to adding your own example to the Gallery.

Browse the Gallery

Please visit this page to browse the Gallery and associated (auto-generated) documentation. There you can also download the scripts and example notebooks for your own use.

A quick summary of the charts currently available in the Gallery is listed below:

Charts in the Repeatable Information Charts Kit (RICK) module

This module was inspired by charts created for the VFH Bylaw Review Report. There was a need to develop a standardized brand and design language for everything BDITTO produces, so this module aims to produce a regularized set of charts and maps that are consistent with previous charts we create. All of the chart/map producing functions returns a matplotlib fig and ax object so that the figure can be furthere modified using matplotlib functions.

geo.ttc(con)

Returns a geopandas dataframe of the TTC subway network.

geo.island(con)

Returns a geopandas dataframe of the Toronto Island.

charts.chloro_map(con, df, lower, upper, title, **kwargs)

This function creates a chloropleth map.

The following arguments must be passed in order for the function to run.

argument variable type description
con SQL Connection Object Used to additional layers from the SQL database.
df GeoPandas DataFrame Data for the chloropleth map. The data must only contain 2 columns; the first column has to be the geom column and the second has to be the data that needs to be mapped.
lower int Lower bound for the colourmap
upper int Upper bound for the colourmap
title str Text string for the title text

Additionally, there are optional arguments that can be passed to the function

argument variable type default description
subway boolean False Flag to plot the subway network on the map. False indicates the subway network does not show up.
island boolean True Flag to plot the Toronto Islands as having no data. True indicates the islands are coloured the same as the Waterfront neighbourhood.
cmap str YlOrRd String to specify colourmap for the map.
unit str None Specifies if a unit should be added to the legend box. The automatic placement of the unit only works if the upper or lower are whole numbers.
nbins int 2 Number of ticks in the colourmap

charts.line_chart(data, ylab, xlab, **kwargs)

Produces a simple line chart. The xaxis is not formatted by this function and requires further manipulation with matplotlib. In addition, annotation boxes must be added on manually with a something like this:

fig.text(0.94, 0.96, '176,000', transform=ax.transAxes, wrap = True, fontsize=9, fontname = 'Libre Franklin',
         verticalalignment='top', ha = 'center', bbox=props, color = purple)

The function defines the styling with the props variable, so the only manipulations needed is the positioning and the text itself.

The following arguments must be passed in order for the function to run.

argument variable type description
data Series or list Data for the chart
ylab str Label for the yaxis
xlab str Label for the xaxis

Additionally, there are optional arguments that can be passed to the function

argument variable type default description
ymin int 0 Lower bound for the yaxis
ymax int The maximum value of the dataset Upper bound for the yaxis
yinc int One-third of the range of the data Interval for the yaxis ticks

charts.tow_chart(data, ylab, **kwargs)

Produces a 7-Day time of week chart that shows data points for each hour during one week. The xaxis is fixed to the 168 hours that produces the week, and the data must be ordered so that the first entry represents Monday at midnight.

The following arguments must be passed in order for the function to run.

argument variable type description
data Series or list Data for the chart
ylab str Label for the yaxis

Additionally, there are optional arguments that can be passed to the function

argument variable type default description
ymin int 0 Lower bound for the yaxis
ymax int The maximum value of the dataset Upper bound for the yaxis
yinc int One-third of the range of the data Interval for the yaxis ticks

Helper functions

function purpose
calculate_params() Finds minimum, maximum and incremental values used to determine the limits and ticks of the dependent axis.
calculate_delta() Required for calculate_params().
init_fig() Initializes the figure object corresponding to the plot being made.
set_plot_style() Sets size, background and grid for plot.
set_grid() Sets the grid for plot. Called by set_plot_style().
set_ticks() Sets x and y axis ticks. Format depends on type of index being used.
set_labels() Set the labels of the y and x axes.
add_bar_annotations() Adds annotations to charts including bars, such as the height of bars or percent changes.
horizontal_bar_annotations() Called by add_bar_annotations() in the case of horizontal graphs.
vertical_bar_annotations() Called by add_bar_annotations() in the case of vertical graphs.
plot_line_data() Used for line-type plots.
plot_grouped_bar_data() Used for bar charts.

Stand-alone examples

WIP.....

Adding to / modifying the Charts class

If you would like to add a new plotting function to the Charts class:

  1. Check through the autogenerated documentation and rick.py to see if there is a similar pre-existing function to avoid duplication.
  2. Try to make use of the helper functions available. If there is functionality that you would like to add that is not covered by the available helper functions, consider updating them or creating new ones. Ideally, we want to avoid repetition of code throughout rick.py to make it easier to debug and add features later on.
  3. Make sure to include docstrings formatted according to the Sphinx gallery requirements (check here) so that they can automatically be added to the autogenerated documentation.
  4. Create some examples that can be added to the online Gallery following the instructions here. Ideally, do not create new examples that depend on database connections so that they are always reproducible.
  5. Create a PR in this repo and make sure to include your new examples. Prior to merging, make sure to manually move the autogenerated contents of sphnix/build/html under docs/.

If you are updating an existing function:

  1. Recreate the corresponding example from the Sphinx gallery and make sure that it still works.
  2. Include this in your PR.

About

A gallery of static Python plots that the Big Data Innovation Team has produced

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published