linking front thumbnails, updating screenshots + pyplot API page #8581

choldgraf · 2017-05-06T06:38:37Z

This links the frontpage thumbnails to the sphinx gallery versions so that we don't have to create them manually. Also moves screenshots.rst to a tutorial so continue the consolidation.

story645 · 2017-05-07T19:57:25Z

:/ categorical doesn't work on numpy < 1.8 (#7988) and I'll go fix that bug now... Nevermind, that's not the bug here...confused then...

QuLogic · 2017-05-07T23:53:21Z

Do you have a screenshot of what it looks like? The main reason for the static images is that they don't look very good when scaled down (not that the existing ones are amazing examples in their own right.)

choldgraf · 2017-05-08T02:25:54Z

@QuLogic see the live version here:

http://predictablynoisy.com/matplotlib/

the only difference I can tell is that the images are slightly smaller, but the resolution is actually higher I think.

@story645 it seems like the error is not related to this PR, no?

story645 · 2017-05-08T03:08:34Z

Yeah, realized it wasn't and edited my comment as such...though did open #8592 to fix that bug since it was so long overdue.

choldgraf · 2017-05-08T03:27:09Z

woot!

story645 · 2017-05-09T03:40:41Z

tutorials/01_introductory/sample_plots.py


+data = np.random.randn(2, 100)


please add seeds to random examples

story645 · 2017-05-09T03:41:53Z

examples/lines_bars_and_markers/categorical_variables.py

+"""
+import matplotlib.pyplot as plt
+
+names = ['group_a', 'group_b', 'group_c']


I'm in favor of more concrete examples. Here's some I did...and yes should have just added to matplotlib:

data = {'apples':10, 'oranges':15, 'lemons':5, 'limes':20}

Also:

cat = ["bored", "happy", "bored", "bored", "happy", "bored"] dog = ["happy", "happy", "happy", "happy", "bored", "happy"] time = ["combing", "drinking","feeding", "napping", "playing", "washing"] fig = plt.figure() ax = fig.add_subplot(1,1,1) ax.plot(time, dog, label="dog") ax.plot(time, cat, label="cat") ax.legend()

lol that's awesome, I'll replace w/ your example here, and maybe leave the tutorial as-is? Or add another example? WDYT?

I think two is good enough since categorical is supported by whatever methods are supported by pyplot and I don't even know that list.

cool - in the latest push I just have two little examples in the example script, and didn't change the tutorial

choldgraf · 2017-05-09T05:50:40Z

ok - comments addressed

choldgraf · 2017-05-09T14:57:36Z

and tests passing!

story645 · 2017-05-09T15:54:43Z

tutorials/01_introductory/pyplot.py

+names = ['group_a', 'group_b', 'group_c']
+values = [1, 10, 100]

+fig, axs = plt.subplots(1, 3, figsize=(9, 3))


this isn't the pyplot interface....

ah yes - you're right...I never use pyplot so do it without thinking :)

choldgraf · 2017-05-09T16:06:12Z

latest push uses pyplot interface

story645 · 2017-05-09T16:17:19Z

tutorials/01_introductory/sample_plots.py


 Here you'll find a host of example plots with the code that
 generated them.

+.. _matplotlibscreenshots:
+
 Simple Plot
 ===========

 Here's a very basic :func:`~matplotlib.pyplot.plot` with text labels:


I like to avoid the word basic/simple/etc. in tutorials. What about "This is how to create a line plot with text labels"

story645 · 2017-05-09T16:22:22Z

tutorials/01_introductory/sample_plots.py

@@ -58,8 +59,8 @@
 You can add arbitrary paths in matplotlib using the


What are arbitrary paths?

not sure - I didn't create any of this text I just moved it here :-P I'll call it "other paths" I guess

nah, just leave it as is then and we can make it someone else's problem. I think the arbitrary is important.

story645 · 2017-05-09T16:26:06Z

tutorials/01_introductory/sample_plots.py

@@ -133,17 +134,17 @@
 Bar charts are simple to create using the :func:`~matplotlib.pyplot.bar`


Same about removing simple. Use :func:~matplotlib.pyplot.bar command to make bar charts, which including customization: (though won't hold you/hold up the pr to this since you didn't write it in the first place)

story645 · 2017-05-09T16:26:51Z

tutorials/01_introductory/sample_plots.py

-.. figure:: ../gallery/pylab_examples/images/sphx_glr_barchart_demo_001.png
-   :target: ../gallery/pylab_examples/barchart_demo.html
+.. figure:: ../../gallery/pylab_examples/images/sphx_glr_barchart_demo_001.png
+   :target: ../../gallery/pylab_examples/barchart_demo.html
   :align: center
   :scale: 50

   Barchart Demo

 It's also simple to create stacked bars


same about simple

choldgraf · 2017-05-09T17:02:19Z

latest push removes the ableist language. The City of Berkeley would be proud of this PR now ;-)

choldgraf · 2017-05-09T23:17:22Z

I think the non-passing-travis build is not related to this PR

choldgraf · 2017-05-09T23:40:25Z

(dunno what's up with the appveyor build...that doesn't build any docs anyway tho I believe)

choldgraf · 2017-05-10T20:02:14Z

can somebody restart those appveyor builds?

QuLogic · 2017-05-10T20:08:37Z

Done.

story645

these are all non-blocking nitpicks

story645 · 2017-05-10T23:08:49Z

doc/api/pyplot_summary.rst

+
+Most of these functions also exist as methods in the
+:class:`matplotlib.axes.Axes` class. You can use them with the
+so-called "Object Oriented" approach to Matplotlib.


don't need so-called (it's redundant)

story645 · 2017-05-10T23:10:15Z

examples/lines_bars_and_markers/categorical_variables.py

+Many times you want to create a plot that uses categorical variables
+in Matplotlib. For example, your data may naturally fall into
+several "bins" and you're interested in summarizing the data per
+bin. Matplotlib allows you to pass categorical variables directly to


Don't think you need this sentence since this is a histogram:

For example, your data may naturally fall into several "bins" and you're interested in summarizing the data per bin.

story645 · 2017-05-10T23:12:56Z

tutorials/01_introductory/sample_plots.py

@@ -243,8 +245,8 @@
 You can plot date data with major and minor ticks and custom tick formatters


You can plot timeseries data ('cause date data is hard to parse)

choldgraf · 2017-05-11T00:15:35Z

ok - I'll hold off to see if anybody else has suggestions, and if not then I'll make @story645 's edits and then merge?

QuLogic

LGTM mostly, but some minor stuff you can fix with @story645's stuff.

QuLogic · 2017-05-11T01:41:45Z

tutorials/01_introductory/pyplot.py

+# Plotting with categorical variables
+# ===================================
+#
+# It is also possible to create a plot using categorical variables


Missing period.

QuLogic · 2017-05-11T01:43:35Z

tutorials/01_introductory/sample_plots.py


-Here's a very basic :func:`~matplotlib.pyplot.plot` with text labels:
+Here's now to create a line plot with  text labels using


Here's how?

QuLogic · 2017-05-11T01:45:20Z

tutorials/01_introductory/sample_plots.py

@@ -55,11 +57,11 @@
 Path demo
 =========

-You can add arbitrary paths in matplotlib using the
+You can add other paths in matplotlib using the


Also needs to be arbitrary cause the arbitrary is in reference to the shape of the part (line, circle, curve)

QuLogic · 2017-05-11T01:47:47Z

examples/lines_bars_and_markers/categorical_variables.py

+Plotting categorical variables
+==============================
+
+How to use categorical variables in matplotlib.


QuLogic · 2017-05-11T01:53:13Z

doc/api/pyplot_summary.rst


 .. currentmodule:: matplotlib.pyplot

 .. autofunction:: plotting

+The Object-Oriented API


Should we put this above the Pyplot API section? It could be formatted as an aside/callout/warning if needed. But I think autofunction generates a big table of functions (your site appears to be rendering a different PR, so I'm not sure), so no-one will ever see this recommendation.

ack yeah I'm sorry - I've got a buncha PRs open so it's hard to remember which one I've pushed to the website. I imagined just adding the section on OO stuff to the pyplot summary. It felt weird to put it at the beginning because the page is called pyplot. What if we instead called the page plotting API instead of pyplot? Then I could add the OO stuff first.

I should note - this is why I put the .. contents:: bit at the top of the page - it'll render a table of contents that will list all the sections. I hoped that would make it easier to find the OO section (and the colormaps section...though TBH I'm still not really sure why that section is on this page ¯\(ツ)/¯ )

Just pushed this PR to the live version: http://predictablynoisy.com/matplotlib/api/pyplot_summary.html

We should probably think about this a bit in a later PR.

choldgraf · 2017-05-11T05:38:34Z

ok comments addressed!

choldgraf · 2017-05-11T17:15:26Z

Looks like all tests are passing - the latest commit was just capitalizing one instance of matplotlib that I had missed...

choldgraf · 2017-05-12T15:32:46Z

anything to do here? ready to 🚢 or still need to ⚓️ ? :D

story645 · 2017-05-12T19:16:16Z

I'm fine with you merging, but lets try pinging @QuLogic

choldgraf · 2017-05-12T21:41:19Z

woo! :)

choldgraf mentioned this pull request May 6, 2017

Final documentation improvements #8555

Closed

11 tasks

dstansby added the Documentation label May 7, 2017

dstansby added this to the 2.1 (next point release) milestone May 7, 2017

choldgraf added 2 commits May 7, 2017 09:00

linking front thumbnails and updating screenshots

0a58484

adding pyplot API context

edf2cf8

choldgraf force-pushed the front_thumbnails branch from 4beaeec to 65fe2ee Compare May 7, 2017 17:15

categorical variable example

aa88fae

choldgraf force-pushed the front_thumbnails branch from 65fe2ee to aa88fae Compare May 7, 2017 17:18

choldgraf changed the title ~~linking front thumbnails and updating screenshots~~ linking front thumbnails, updating screenshots + API pages May 7, 2017

choldgraf changed the title ~~linking front thumbnails, updating screenshots + API pages~~ linking front thumbnails, updating screenshots + pyplot API page May 7, 2017

story645 reviewed May 9, 2017

View reviewed changes

addressing comments

a46410e

choldgraf force-pushed the front_thumbnails branch from 0347764 to a46410e Compare May 9, 2017 16:04

story645 reviewed May 9, 2017

View reviewed changes

ableist language

166a431

story645 approved these changes May 10, 2017

View reviewed changes

QuLogic reviewed May 11, 2017

View reviewed changes

addressing comments

7e5bee9

Update categorical_variables.py

cf76980

QuLogic merged commit 9a80cf6 into matplotlib:master May 12, 2017

		@@ -58,8 +59,8 @@
		You can add arbitrary paths in matplotlib using the

		@@ -133,17 +134,17 @@
		Bar charts are simple to create using the :func:`~matplotlib.pyplot.bar`

		@@ -243,8 +245,8 @@
		You can plot date data with major and minor ticks and custom tick formatters


		Here's a very basic :func:`~matplotlib.pyplot.plot` with text labels:
		Here's now to create a line plot with text labels using

Uh oh!

linking front thumbnails, updating screenshots + pyplot API page #8581

linking front thumbnails, updating screenshots + pyplot API page #8581

Uh oh!

Conversation

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!