Letter Value Plots #661

drewokane · 2015-07-30T19:53:01Z

This pull request adds the _LVPlotter class, lettervalueplot function, documentation, and usage examples.

TODO

Added letter value plot class and plotting function. In addition, added documentation and examples.

mwaskom · 2015-07-30T20:55:10Z

If you want to work out the pep8 errors locally you can install pip install pep8==1.5 which will be the version on travis. The pep8 command got a little stricter and some things in seaborn fail, so just running the latest version won't be that informative about the new code.

mwaskom · 2015-07-30T20:55:41Z

I think the function could be called lvplot like it is in R. A little confusing with lmplot, but lettervalueplot is such a mouthful.

mwaskom · 2015-07-30T20:56:44Z

seaborn/categorical.py

+                 orient, color, palette, saturation,
+                 width, k_depth, linewidth, box_widths):
+
+        if width is None:


These default values should be specified in the function signature for lvplot.

mwaskom · 2015-07-30T21:00:18Z

This looks good on a very brief first pass, but it will be very important to get tests on the numerical internals.

mwaskom · 2015-07-30T21:01:51Z

seaborn/categorical.py

+                ax.plot([y, y], [x -  w / 2, x + w / 2], c='k', alpha=.45, **kws)
+
+                ax.scatter(outliers, np.repeat(x, len(outliers)),
+                           marker=r"$\ast$", c=color)


Why not just a normal "o"?

I didn't really have a preference as far as marker for outliers. Matplotlib uses '+' markers, so I could try to be consistent with Matplotlib or go with 'o'. Thoughts?

I liked the circles you had in the example plots I just think it will be faster to draw those with "o" than with a custom latex marker, right?

Or you could just use "d" to be consistent with seaborn boxplots i guess...

Changed it to 'd' for consistency.

mwaskom · 2015-07-31T04:02:18Z

Also before I forget this should get added as a plot kind in factorplot too eventually.

mwaskom · 2015-07-31T04:03:32Z

The median lines extend past the width of the boxes, so they overlap on nested plots. Is this intentional? I don't love the effect.

mwaskom · 2015-07-31T04:08:01Z

I find this kind of weird:

df = pd.DataFrame({"normal": np.random.normal(3, 1, 10000),
                   "gamma": np.random.gamma(3, .5, 10000)})

sns.lettervalueplot(data=df)
sns.violinplot(data=df, scale="width")

I think I still don't quite understand how these things get drawn, but I would have expected the letter-value plots to look more like a discretized violinplot. Is that wrong?

drewokane · 2015-07-31T17:26:11Z

Your intuition wouldn't be off in this case. The widths of the boxes are determined by which width function is chosen in _lvplot. My initial stab at this I liked the look of the "linear" widths, but if you use the "exponential" widths it actually looks like a discrete violin plot (as you mentioned). Perhaps changing the default to "exponential" will be better for consistency sake. (Also, I realized there is a bug, in that I am not passing the keyword argument to the plotter (embarassed))

df = pd.DataFrame({"normal": np.random.normal(3, 1, 10000),
                   "gamma": np.random.gamma(3, .5, 10000)})

sns.violinplot(data=df, scale="width")
sns.lettervalueplot(data=df, box_widths='exponential')

mwaskom · 2015-07-31T17:27:24Z

Ah, OK. I had played around with that keyword argument, but couldn't seem to get the effect I was looking for. I guess that explains why :)

mwaskom · 2015-07-31T17:43:21Z

Speaking of, I think the box_widths parameter might make more sense as scale, for better consistency with the violinplot paramter. The values aren't the same, but they have a similar effect. Also for some reason I kept expecting box_widths to take a numeric size value.

mwaskom · 2015-08-28T02:38:13Z

Any update on tests?

drewokane · 2015-08-28T17:12:52Z

I've added some basic tests so far and plan on adding tests for the numerical internals, e.g. testing the letter value calculations, etc. I'll push the latest edits I have today.

mwaskom · 2015-08-28T17:17:57Z

Cool. By the way, I would generally advise writing the tests in the other direction. I've found that it's best to start directly testing the core numeric parts, then to pull back and test some of the plotting methods, and finally pull back further and add some high-level smoke tests for the function itself (what I assume you mean by "basic tests"). The reason is that this way lets the code coverage report be a useful guide to how complete the tests are. If you start with tests that run the function and just ask "are there some plots on the axes", it will look like the internals are "covered", but the plots very well may be wrong. It might be helpful to rename the basic tests so nose doesn't run them as you develop and use the coverage report to make sure you're really finding all the corners in the numeric code. Anyway, just some thoughts.

mwaskom · 2015-08-28T17:18:20Z

Also can you add a TODO list in the original issue just so we can track the items that have come up in the comments?

…hes, not artists.

…ange [0, 1].

drewokane · 2015-09-12T19:58:04Z

I've completed writing the tests (making them Python 3 forward compatible) and the other tasks on the checklist. I've checked the tests (locally with nosetests) and the only failures I can see are the ones connected with other modules, not TestLVPlotter.

mwaskom · 2015-09-12T20:03:48Z

The Python 3 errors are in TestLVPlotter. It's hard to tell with the Python 2 build as it didn't make it past the PEP8 check.

drewokane · 2015-09-12T20:29:18Z

Yeah I checked Travis. I didn't realize filter and map were iterators in Python 3. Hopefully the modifications pass muster.

drewokane · 2015-09-12T20:48:03Z

Unfortunately I can't duplicate the PEP8 error Travis is getting (and I made sure to install pep8==1.5).

mwaskom · 2015-09-12T20:51:38Z

seaborn/tests/test_categorical.py

+    def test_draw_missing_boxes(self):
+
+        ax = cat.lvplot("g", "y", data=self.df,
+                         order=["a", "b", "c", "d"])


This line is over-indented

drewokane · 2015-09-12T22:05:37Z

Build is passing on all versions, PEP8 is all done.

drewokane · 2015-10-03T18:19:29Z

Any more comments, concerns, suggestions?

drewokane · 2015-11-14T21:28:16Z

Not trying to be bothersome but I wanted to check in a see if you had any further comments or suggestions for the PR? I completely understand if you're swamped with other stuff.

mwaskom · 2015-11-15T01:09:52Z

Yeah, sorry about letting this drop. This is a substantial enough addition that I've really wanted to make time to go through it in detail and make sure I understand how everything works. Unfortunately, I've been very busy the past few months (and will be for the foreseeable future) and just have not been able to devote a large enough block of time to get my head around everything here. But since you put so much effort into this, I really do feel bad about letting this hang in limbo. I'll merge it now so that it can start to get some real use, hopefully any residual issues will get identified and ironed out before the next release. If you could try to stay on top of anything that pops up, that would be very helpful.

Letter Value Plots

stonebig · 2015-11-15T10:22:35Z

Hi,

May the documentation take Matplotlib 1.5 into account now ?

ax = sns.lvplot(x="size", y="tip", data=tips.sort("size"))
# or, if you are on Matplotlib 1.5+
ax = sns.lvplot(x="size", y="tip", data=tips.sort_values(by="size"))

drewokane · 2015-11-15T19:09:31Z

@mwaskom Sure, I can stay on top of any issues that pop up. Thanks!

mwaskom · 2015-11-15T20:23:42Z

May the documentation take Matplotlib 1.5 into account now ?

This is a change in pandas, not matplotlib, but yes that example should be updated.

drewokane added 3 commits July 30, 2015 11:44

Add letter value plot.

b3eed79

Added letter value plot class and plotting function. In addition, added documentation and examples.

Updated letter value plot docs.

c3e368d

Spelling fix.

e21840e

mwaskom reviewed Jul 30, 2015
View reviewed changes

Drew O'Kane added 3 commits July 31, 2015 10:30

Fixed bug with box_widths. Wasn't passing keyword to _lvplot.

3b057dc

Made median bar same width as boxes.

567b276

Changed outliers to d. More consistent with Seaborn boxplot.

7017de1

Drew O'Kane added 4 commits August 1, 2015 12:01

Changed p to outlier_prop. Minor fixes.

006ef5d

Changed name of plotting function to lvplot.

238b617

PEP8 fixes.

9d6cf40

Fixed division by zero error.

f451e00

Added numerical tests.

840789d

lukauskas mentioned this pull request Sep 3, 2015

feature request: violinplot like plot using histograms instead of kde #688

Closed

Drew O'Kane added 6 commits September 3, 2015 11:14

Changed from xrange to range for future compatability.

5df9287

Fixed single observation plotting bug. Fixed tests for number of patc…

ad096d1

…hes, not artists.

Finished letter value tests.

9cf9159

Added outlier_prop default value, checking for value in appropriate r…

db7b10e

…ange [0, 1].

Docstring for outlier_prop

500f7d8

Fixed tests.

b8ae90e

Fixed tests so Python 3 compatible.

79add92

mwaskom reviewed Sep 12, 2015
View reviewed changes

Drew O'Kane added 3 commits September 12, 2015 13:52

Python 3 fixes.

31854be

More Python 3 fixes.

931b8bc

More Python 3 fixes. Removed itertools references.

42046c6

mwaskom added a commit that referenced this pull request Nov 15, 2015

Merge pull request #661 from drewokane/lv_plot

1078424

Letter Value Plots

mwaskom merged commit 1078424 into mwaskom:master Nov 15, 2015

drewokane deleted the lv_plot branch November 15, 2015 19:10

MaozGelbart mentioned this pull request Aug 29, 2020

boxenplot area scale calculation #2232

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Letter Value Plots #661

Letter Value Plots #661

drewokane commented Jul 30, 2015

mwaskom commented Jul 30, 2015

mwaskom commented Jul 30, 2015

mwaskom Jul 30, 2015

mwaskom commented Jul 30, 2015

mwaskom Jul 30, 2015

drewokane Jul 30, 2015

mwaskom Jul 30, 2015

mwaskom Jul 30, 2015

drewokane Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Jul 31, 2015

drewokane commented Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Aug 28, 2015

drewokane commented Aug 28, 2015

mwaskom commented Aug 28, 2015

mwaskom commented Aug 28, 2015

drewokane commented Sep 12, 2015

mwaskom commented Sep 12, 2015

drewokane commented Sep 12, 2015

drewokane commented Sep 12, 2015

mwaskom Sep 12, 2015

drewokane commented Sep 12, 2015

drewokane commented Oct 3, 2015

drewokane commented Nov 14, 2015

mwaskom commented Nov 15, 2015

stonebig commented Nov 15, 2015

drewokane commented Nov 15, 2015

mwaskom commented Nov 15, 2015

Letter Value Plots #661

Letter Value Plots #661

Conversation

drewokane commented Jul 30, 2015

TODO

mwaskom commented Jul 30, 2015

mwaskom commented Jul 30, 2015

mwaskom Jul 30, 2015

Choose a reason for hiding this comment

mwaskom commented Jul 30, 2015

mwaskom Jul 30, 2015

Choose a reason for hiding this comment

drewokane Jul 30, 2015

Choose a reason for hiding this comment

mwaskom Jul 30, 2015

Choose a reason for hiding this comment

mwaskom Jul 30, 2015

Choose a reason for hiding this comment

drewokane Jul 31, 2015

Choose a reason for hiding this comment

mwaskom commented Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Jul 31, 2015

drewokane commented Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Jul 31, 2015

mwaskom commented Aug 28, 2015

drewokane commented Aug 28, 2015

mwaskom commented Aug 28, 2015

mwaskom commented Aug 28, 2015

drewokane commented Sep 12, 2015

mwaskom commented Sep 12, 2015

drewokane commented Sep 12, 2015

drewokane commented Sep 12, 2015

mwaskom Sep 12, 2015

Choose a reason for hiding this comment

drewokane commented Sep 12, 2015

drewokane commented Oct 3, 2015

drewokane commented Nov 14, 2015

mwaskom commented Nov 15, 2015

stonebig commented Nov 15, 2015

drewokane commented Nov 15, 2015

mwaskom commented Nov 15, 2015