Skip to content

Draft version of MEP28: Simplification of boxplots #7282

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Oct 19, 2016
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
326 changes: 326 additions & 0 deletions doc/devel/MEP/MEP28.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,326 @@
=============================================
MEP 28: Remove Complexity from Axes.boxplot
=============================================

.. contents::
:local:

Status
======

..
.. MEPs go through a number of phases in their lifetime:

- **Discussion**
..
.. - **Progress**: Consensus was reached on the mailing list and
.. implementation work has begun.
..
.. - **Completed**: The implementation has been merged into master.
..
.. - **Superseded**: This MEP has been abandoned in favor of another
.. approach.

Branches and Pull requests
==========================

Adding pre- & post-processing options to ``cbook.boxplot_stats``: https://github.com/phobson/matplotlib/tree/boxplot-stat-transforms
Exposing ``cbook.boxplot_stats`` through ``Axes.boxplot`` kwargs: None
Remove redundant statistical kwargs in ``Axes.boxplot``: None
Remove redundant style options in ``Axes.boxplot``: None
Remaining items that arise through discussion: None

Abstract
========

Over the past few releases, the ``Axes.boxplot`` method has grown in
complexity to support fully customizable artist styling and statistical
computation. This lead to ``Axes.boxplot`` being split off into multiple
parts. The statistics needed to draw a boxplot are computed in
``cbook.boxplot_stats``, while the actual artists are drawn by ``Axes.bxp``.
The original method, ``Axes.boxplot`` remains as the most public API that
handles passing the user-supplied data to ``cbook.boxplot_stats``, feeding
the results to ``Axes.bxp``, and pre-processing style information for
each facet of the boxplot plots.

This MEP will outline a path forward to rollback the added complexity
and simplify the API while maintaining reasonable backwards
compatibility.

Detailed description
====================

Currently, the ``Axes.boxplot`` method accepts parameters that allow the
users to specify medians and confidence intervals for each box that
will be drawn in the plot. These were provided so that avdanced users
could provide statistics computed in a different fashion that the simple
method provided by matplotlib. However, handling this input requires
complex logic to make sure that the forms of the data structure match what
needs to be drawn. At the moment, that logic contains 9 separate if/else
statements nested up to 5 levels deep with a for loop, and may raise up to 2 errors.
These parameters were added prior to the creation of the ``Axes.bxp`` method,
which draws boxplots from a list of dictionaries containing the relevant
statistics. Matplotlib also provides a function that computes these
statistics via ``cbook.boxplot_stats``. Note that advanced users can now
either a) write their own function to compute the stats required by
``Axes.bxp``, or b) modify the output returned by ``cbook.boxplots_stats``
to fully customize the position of the artists of the plots. With this
flexibility, the parameters to manually specify only the medians and their
confidences intervals remain for backwards compatibility.

Around the same time that the two roles of ``Axes.boxplot`` were split into
``cbook.boxplot_stats`` for computation and ``Axes.bxp`` for drawing, both
``Axes.boxplot`` and ``Axes.bxp`` were written to accept parameters that
individually toggle the drawing of all components of the boxplots, and
parameters that individually configure the style of those artists. However,
to maintain backwards compatibility, the ``sym`` parameter (previously used
to specify the symbol of the fliers) was retained. This parameter itself
requires fairly complex logic to reconcile the ``sym`` parameters with the
newer ``flierprops`` parameter at the default style specified by ``matplotlibrc``.

This MEP seeks to dramatically simplify the creation of boxplots for
novice and advanced users alike. Importantly, the changes proposed here
will also be available to downstream packages like seaborn, as seaborn
smartly allows users to pass arbitrary dictionaries of parameters through
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So, a little unclear about scope delineation here. What does seaborn do above and beyond mpl? And how is this not creeping into seaborn's territory.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

My thought is that these few changes should be implemented in such a way that seaborn users have access to them without seaborn needing to change at all. Shooting for simplicity and flexibility for users and downstream libraries

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, so you want to maintain downstream use, but you plan to remove **kwargs. Is there verification that seaborn doesn't use those args? Granted, this seems like it'd be a pretty small PR on seaborn if it does change things with them..

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

seaborn uses very few parameters and just passes everything else from the user as **kws. Also instances of call matplotlib's boxplot API are below:

  1. https://github.com/mwaskom/seaborn/blob/dd8ab4985eadba43c94542cfad56a78dcfefef31/seaborn/categorical.py#L468
  2. https://github.com/mwaskom/seaborn/blob/dd8ab4985eadba43c94542cfad56a78dcfefef31/seaborn/categorical.py#L497

Next step would be to checkout yhat's python-ggplot

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In some ways that's probably good in that it means you don't have to worry about downstream support with ggplot.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

you're absolutely right about that. i guess it just makes me feel insecure about our implementation.

the seaborn API to the underlying matplotlib functions.

This will be achieved in the following way:

1. ``cbook.boxplot_stats`` will be modified to allow pre- and post-
computation transformation functions to be passed in (e.g., ``np.log``
and ``np.exp`` for lognormally distributed data)
2. ``Axes.boxplot`` will be modified to also accept and naïvely pass them
to ``cbook.boxplots_stats`` (Alt: pass the stat function and a dict
of its optional parameters).
3. Outdated parameters from ``Axes.boxplot`` will be deprecated and
later removed.

Implementation
==============

Passing transform functions to ``cbook.boxplots_stats``
-------------------------------------------------------

This MEP proposes that two parameters (e.g., ``transform_in`` and
``transform_out`` be added to the cookbook function that computes the
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm thinking pre_transform and post_transform instead, though not for any particular reason.

statistics for the boxplot function. These will be optional keyword-only
arguments and can easily be set to ``lambda x: x`` as a no-op when omitted
by the user. The ``transform_in`` function will be applied to the data
as the ``boxplot_stats`` function loops through each subset of the data
passed to it. After the list of statistics dictionaries are computed the
``transform_out`` function is applied to each value in the dictionaries.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wouldn't it make more sense to apply it to the entire dictionary? All the individual things don't make much sense on their own.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we're saying the same thing, but AFAIK, you can't pass a dictionary to e.g., np.exp. So you have to loop through the items. That's all I mean.

for key, value in stat_dict.items():
    if key != 'label':
        stat_dict[key] = transform_out(value)


These transformations can then be added to the call signature of
``Axes.boxplot`` with little impact to that method's complexity. This is
because they can be directly passed to ``cbook.boxplot_stats``.
Alternatively, ``Axes.boxplot`` could be modified to accept an optional
statistical function kwarg and a dictionary of parameters to be direcly
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't quite understand this alternative.

passed to it.

At this point in the implementation users and external libraries like
seaborn would have complete control via the ``Axes.boxplot`` method. More
importantly, at the very least, seaborn would require no changes to its
API to allow users to take advantage of these new options.

Simplifications to the ``Axes.boxplot`` API and other functions
---------------------------------------------------------------

Simplifying the boxplot method consists primarily of deprecating and then
removing the redundant parameters. Optionally, a next step would include
rectifying minor terminological inconsistencies between ``Axes.boxplot``
and ``Axes.bxp``.

The parameters to be deprecated and removed include:

1. ``usermedians`` - processed by 10 SLOC, 3 ``if`` blocks, a ``for`` loop
2. ``conf_intervals`` - handled by 15 SLOC, 6 ``if`` blocks, a ``for`` loop
3. ``sym`` - processed by 12 SLOC, 4 ``if`` blocks

Removing the ``sym`` option allows all code in handling the remaining
styling parameters to be moved to ``Axes.bxp``. This doesn't remove
any complexity, but does reinforce the single responsibility principle
among ``Axes.bxp``, ``cbook.boxplot_stats``, and ``Axes.boxplot``.

Additionally, the ``notch`` parameter could be renamed ``shownotches``
to be consistent with ``Axes.bxp``. This kind of cleanup could be taken
a step further and the ``whis``, ``bootstrap``, ``autorange`` could
be rolled into the kwargs passed to the new ``statfxn`` parameter.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What is the statfxn parameter? It hasn't been mentioned before.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'll clarify this in the MEP but I'll explain here too. From my POV, there are two options available. The first looks like this:

def boxplot_stats(data, ..., transform_in=None, transform_out=None):
    if transform_in is None:
        transform_in = lambda x: x

    if transform_out is None:
        transform_out = lambda x: x

    output = []
    for _d in data:
        d = transform_in(_d)
        stat_dict = do_stats(d)
        for key, value in stat_dict.item():
            if key != 'label':
                stat_dict[key] = transform_out(value)
        output.append(d)
    return output

class Axes(...):
    def boxplot_option1(data, ..., transform_in=None, transform_out=None):
        stats = cbook.boxplot_stats(data, ..., transform_in=transform_in, transform_out=transform_out)
        return self.bxp(stats, ...)

    def boxplot_option2(data, ..., statfxn=None, **statopts):
        if statfxn is None:
            statfxn = boxplot_stats
        stats = statfxn(data, **statopts)
        return self.bxp(stats, ...)

So in both cases you can do:

fig, ax1 = plt.subplots()
artists1 = ax1.boxplot_optionX(data, transform_in=np.log, transform_out=np.exp)

But Option two lets a user write a completely custom stat function (my_box_stats) with fancy BCA confidence intervals and the whiskers set differently depending on some attribute of the data.

Currently, users can do:

fig, ax1 = plt.subplots()
my_stats = my_box_stats(data,  bootstrap_method='BCA', whisker_method='dynamic')
ax1.bxp(my_stats)

Under Option 2, the user could do (only slightly more convenient):

fig, ax1 = plt.subplots()
ax1.boxplot(data, statfxn=my_box_stats, bootstrap_method='BCA', whisker_method='dynamic')

It's not that compelling of a difference, I admit. But Option 2 would let the following be valid in seaborn:

sns.factorplot(x="day", y="total_bill", hue="sex", data=tips, kind='box',
    palette="PRGn", shownotches=True, statfxn=my_box_stats, 
    bootstrap_method='BCA', whisker_method='dynamic')

So a lot of downstream flexibility is gained (in both approaches) with little implementation.


Backward compatibility
======================

Implementation of this MEP would eventually result in the backwards
incompatible deprecation and then removal of the keyword parameters
``usermedians``, ``conf_intervals``, and ``sym``. Cursory searches on
GitHub indicated that ``usermedians``, ``conf_intervals`` are used by
few users, who all seem to have a very strong knowledge of matplotlib.
A robust deprecation cycle should provide sufficient time for these
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Considering we are have discussion about one release cycle versus two release cycle on the finance module right now, it might be worth being specific here.

users to migrate to a new API.

Deprecation of ``sym`` however, may have a much broader reach into
the matplotlib userbase.

Schedule
--------
An accelerated timeline could look like the following:

#. v2.0.1 add transforms to ``cbook.boxplots_stats``, expose in ``Axes.boxplot``
#. v2.1.0 deprecate ``usermedians``, ``conf_intervals``, ``sym`` parameters
#. v2.2.0 make deprecations noisier
#. v2.3.0 remove ``usermedians``, ``conf_intervals``, ``sym`` parameters
#. v2.3.0 deprecate ``notch`` in favor of ``shownotches`` to be consistent with other parameters and ``Axes.bxp``
#. v2.4.0 remove ``notch`` parameter, move all style and artist toggling logic to ``Axes.bxp``. ``Axes.boxplot`` is little more than a broker between ``Axes.bxp`` and ``cbook.boxplots_stats``


Anticipated Impacts to Users
----------------------------

As described above deprecating ``usermedians`` and ``conf_intervals``
will likely impact few users. Those who will be impacted are almost
certainly advanced users who will be able to adapt to the change.

Deprecating the ``sym`` option may import more users and effort should
be taken to collect community feedback on this.

Anticipated Impacts to Downstream Libraries
-------------------------------------------

The source code (GitHub master as of 2016-10-17) was inspected for
seaborn and python-ggplot to see if these changes would impact their
use. None of the parameters nominated for removal in this MEP are used by
seaborn. The seaborn APIs that use matplotlib's boxplot function allow
user's to pass arbitrary ``**kwargs`` through to matplotlib's API. Thus
seaborn users with modern matplotlib installations will be able to take
full advantage of any new features added as a result of this MEP.

Python-ggplot has implemented its own function to draw boxplots. Therefore,
no impact can come to it as a result of implementing this MEP.

Alternatives
============

Variations on the theme
-----------------------

This MEP can be divided into a few loosely coupled components:

#. Allowing pre- and post-computation tranformation function in ``cbook.boxplot_stats``
#. Exposing that transformation in the ``Axes.boxplot`` API
#. Removing redundant statistical options in ``Axes.boxplot``
#. Shifting all styling parameter processing from ``Axes.boxplot`` to ``Axes.bxp``.


With this approach, #2 depends and #1, and #4 depends on #3.

There are two possible approaches to #2. The first and most direct would
be to mirror the new ``transform_in`` and ``tranform_out`` parameters of
``cbook.boxplot_stats`` in ``Axes.boxplot`` and pass them directly.

The second approach would be to add ``statfxn`` and ``statfxn_args``
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

any chance of psuedocode like examples of the two approaches?

parameters to ``Axes.boxplot``. Under this implementation, the default
value of ``statfxn`` would be ``cbook.boxplot_stats``, but users could
pass their own function. Then ``transform_in`` and ``tranform_out`` would
then be passed as elements of the ``statfxn_args`` parameter.

.. python:
def boxplot_stats(data, ..., transform_in=None, transform_out=None):
if transform_in is None:
transform_in = lambda x: x

if transform_out is None:
transform_out = lambda x: x

output = []
for _d in data:
d = transform_in(_d)
stat_dict = do_stats(d)
for key, value in stat_dict.item():
if key != 'label':
stat_dict[key] = transform_out(value)
output.append(d)
return output


class Axes(...):
def boxplot_option1(data, ..., transform_in=None, transform_out=None):
stats = cbook.boxplot_stats(data, ...,
transform_in=transform_in,
transform_out=transform_out)
return self.bxp(stats, ...)

def boxplot_option2(data, ..., statfxn=None, **statopts):
if statfxn is None:
statfxn = boxplot_stats
stats = statfxn(data, **statopts)
return self.bxp(stats, ...)

Both cases would allow users to do the following:

.. python:
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To be very annoying, ( 😄 ) could you add an example of how you would do this with the current API?
The difference would be more striking.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@NelleV great suggestion. done.

fig, ax1 = plt.subplots()
artists1 = ax1.boxplot_optionX(data, transform_in=np.log,
transform_out=np.exp)


But Option Two lets a user write a completely custom stat function
(e.g., ``my_box_stats``) with fancy BCA confidence intervals and the
whiskers set differently depending on some attribute of the data.

This is available under the current API:

.. python:
fig, ax1 = plt.subplots()
my_stats = my_box_stats(data, bootstrap_method='BCA',
whisker_method='dynamic')
ax1.bxp(my_stats)

And would be more concise with Option Two

.. python:
fig, ax = plt.subplots()
statopts = dict(transform_in=np.log, transform_out=np.exp)
ax.boxplot(data, ..., **statopts)

Users could also pass their own function to compute the stats:

.. python:
fig, ax1 = plt.subplots()
ax1.boxplot(data, statfxn=my_box_stats, bootstrap_method='BCA',
whisker_method='dynamic')

From the examples above, Option Two seems to have only marginal benifit,
but in the context of downstream libraries like seaborn, its advantage
is more apparent as the following would be possible without any patches
to seaborn:

.. python:
import seaborn
tips = seaborn.load_data('tips')
g = seaborn.factorplot(x="day", y="total_bill", hue="sex", data=tips,
kind='box', palette="PRGn", shownotches=True,
statfxn=my_box_stats, bootstrap_method='BCA',
whisker_method='dynamic')

This type of flexibility was the intention behind splitting the overall
boxplot API in the current three functions. In practice however, downstream
libraries like seaborn support versions of matplotlib dating back well
before the split. Thus, adding just a bit more flexibility to the
``Axes.boxplot`` could expose all the functionality to users of the
downstream libraries with modern matplotlib installation without intervention
from the downstream library maintainers.

Doing less
----------

Another obvious alternative would be to omit the added pre- and post-
computation transform functionality in ``cbook.boxplot_stats`` and
``Axes.boxplot``, and simply remove the redundant statistical and style
parameters as described above.

Doing nothing
-------------

As with many things in life, doing nothing is an option here. This means
we simply advocate for users and downstream libraries to take advantage
of the split between ``cbook.boxplot_stats`` and ``Axes.bxp`` and let
them decide how to provide an interface to that.
1 change: 1 addition & 0 deletions doc/devel/MEP/index.rst
Original file line number Diff line number Diff line change
Expand Up @@ -29,3 +29,4 @@ Matplotlib Enhancement Proposals
MEP25
MEP26
MEP27
MEP28