Fix Scatter plot datetime and Axis #12949

nparley · 2016-04-21T17:34:39Z

closes #10611
closes #8113
closes #10678

tests added / passed
passes git diff upstream/master | flake8 --diff
whatsnew entry

This is a first go at trying to fix the above problems. I have created two before and after notebooks here: https://gist.github.com/nparley/08c81672b0e3d7d85b00694a671f57ad

jreback · 2016-04-21T17:55:40Z

@TomAugspurger @sinhrks

sinhrks · 2016-04-21T21:52:50Z

pandas/tools/plotting.py

-    def __init__(self, data, x, y, **kwargs):
-        MPLPlot.__init__(self, data, **kwargs)
+    def __init__(self, data, x, y, sharex=False, **kwargs):
+        if sharex is None:


To apply the same fix to hexbin, pls move the logic to PlanePlot. And I think sharex should be always False because we don't support scatter subplots ATM?

@sinhrks This is in PlanePlot

jreback · 2016-05-20T13:32:28Z

can you rebase / update

codecov-io · 2016-05-20T21:54:48Z

Codecov Report

Merging #12949 into master will decrease coverage by <.01%.
The diff coverage is 73.07%.

@@            Coverage Diff             @@
##           master   #12949      +/-   ##
==========================================
- Coverage      91%      91%   -0.01%     
==========================================
  Files         145      145              
  Lines       49576    49605      +29     
==========================================
+ Hits        45119    45142      +23     
- Misses       4457     4463       +6

Flag	Coverage Δ
#multiple	`88.77% <73.07%> (-0.01%)`	⬇️
#single	`40.56% <15.38%> (-0.02%)`	⬇️

Impacted Files	Coverage Δ
pandas/core/internals.py	`93.55% <100%> (ø)`	⬆️
pandas/core/generic.py	`96.26% <100%> (ø)`	⬆️
pandas/tools/plotting.py	`71.72% <66.66%> (-0.04%)`	⬇️
pandas/core/categorical.py	`95.85% <0%> (ø)`	⬆️
pandas/core/base.py	`95.51% <0%> (ø)`	⬆️
pandas/core/series.py	`94.89% <0%> (ø)`	⬆️
pandas/core/algorithms.py	`94.57% <0%> (+0.04%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 56c2019...cbf3c73. Read the comment docs.

jorisvandenbossche · 2016-08-13T19:40:29Z

@TomAugspurger @sinhrks What's the status of this?

sinhrks · 2016-08-13T21:39:15Z

@nparley can u update to fix above points and add tests?

nparley · 2016-08-13T21:54:25Z

Yes sorry I have a go at bringing this up to date

nparley · 2016-08-15T10:32:12Z

@sinhrks I have moved the login into the PlanePlot class. What stops this working for hexbin is that I have only overridded _compute_plot_data in ScatterPlot and not in PlanePlot. This is because hexbin seems to overflow if you just send it a date. To get hexbin working with dates would need the dates to be converted to numbers and then the axis formatted.

nparley · 2016-08-16T12:59:24Z

This is actually passing I think there was a bit of a travis problem with the latest run (https://travis-ci.org/nparley/pandas/builds/152366202)

jreback · 2016-09-09T22:36:03Z

can you rebase / update?

nparley · 2016-09-10T13:00:35Z

@jreback this seems to be failing the lint but not in a file related to this PR https://travis-ci.org/nparley/pandas/jobs/158939635 that I can see?

Maybe #14198 will fix the lint error.

jreback · 2016-09-10T13:59:30Z

flake is too strict here - maybe the version changed or something

it's not comparing case

nparley · 2016-09-10T14:36:26Z

Line 685 and 690 are both key '2000q4' it's line 681 which is the different case '2000Q4'.

jreback · 2016-09-10T14:40:36Z

@nparley hmm, you are right. but it doesn't fail flaking. can you get it to fail on travis (and then fix)?

jreback · 2016-09-10T14:40:57Z

prob need to specify that option in setup.cfg to make it active

nparley · 2016-09-10T14:43:29Z

Taking out the duplicate line as I did in that #14198 seems to pass locally and on travis. I installed a clean flake8 about 2 hours a go. This was linting locally on my PC while doing the #14198.

linting -> pandas/tseries
pandas/tseries/tests/test_tslib.py:685:18: F999 dictionary key '2000q4' repeated with different values
pandas/tseries/tests/test_tslib.py:690:18: F999 dictionary key '2000q4' repeated with different values
linting -> pandas/tests

to

linting -> pandas/compat
linting -> pandas/sparse
linting -> pandas/tools
linting -> pandas/tseries
linting -> pandas/tests
linting -> pandas/computation
linting -> pandas/util

jreback · 2016-09-10T14:54:16Z

its doesn't fail on travis, so not sure what the difference is.

nparley · 2016-09-10T15:14:22Z

It did not pass on travis for me. This is the current master run on my own travis 2 hours a go: https://travis-ci.org/nparley/pandas/builds/158958462 and this is with that duplicate line removed https://travis-ci.org/nparley/pandas/builds/158959215

jreback · 2016-09-10T15:22:13Z

https://travis-ci.org/pydata/pandas/builds/158970327 here is master, just passed.

make sure you are rebased.

nparley · 2016-09-10T15:40:18Z

Rebased hopefully

jreback · 2016-11-22T11:46:02Z

status of this?

nparley · 2016-11-22T11:55:59Z

I think it's awaiting feedback to see if anything else needs to be added. I updated the PR after the first comments. I.e. I don't think it's waiting for anything my end.

jorisvandenbossche · 2016-11-22T18:58:30Z

@nparley can you add tests for each of the issues that this would close?

jreback · 2017-01-21T23:15:07Z

@nparley can you update and respond to comments?

jreback · 2017-03-20T13:54:41Z

@nparley can you rebase / update

jorisvandenbossche

@nparley Can you also add some tests for this?

jorisvandenbossche · 2017-03-20T22:59:04Z

pandas/core/internals.py

@@ -3289,6 +3289,16 @@ def get_numeric_data(self, copy=False):
        self._consolidate_inplace()
        return self.combine([b for b in self.blocks if b.is_numeric], copy)

+    def get_dt_data(self, copy=False):


I would call this get_datetime_data

jorisvandenbossche · 2017-03-20T23:09:46Z

pandas/tools/plotting.py

@@ -1574,6 +1577,31 @@ def _post_plot_logic(self, ax, data):
        ax.set_ylabel(pprint_thing(y))
        ax.set_xlabel(pprint_thing(x))

+    def _compute_plot_data_with_dt(self):


If this is only called in _compute_plot_data in ScatterPlot, why not just putting the code in that function?

jorisvandenbossche · 2017-03-20T23:09:58Z

pandas/tools/plotting.py

+    def __init__(self, data, x, y, sharex=False, **kwargs):
+        if sharex is None:
+            # Fix x axis color bar problems
+            sharex = False


This sharex fix, is this independent from the datetime fix?
If so, can you make a separate PR for it?

… (i.e. scatter) is happy with datetimes

nparley · 2017-04-09T15:31:44Z

Hi Guys, I think this PR is probably quite stale now. My development work is not touching pandas that much unfortunately and I don't really have that much time currently. Maybe some of the ideas can be moved into a more active PR if they are still applicable? I have tired to rebase and address the comments above to clean things up but have not tested the latest commit.

jreback · 2017-05-13T21:36:26Z

would need a rebase

jreback · 2017-08-01T22:55:52Z

closing as stale. if you want to re-visit, pls comment and we can re-open.

jreback added Bug Visualization plotting labels Apr 21, 2016

sinhrks reviewed Apr 21, 2016
View reviewed changes

jorisvandenbossche reviewed Mar 20, 2017

View reviewed changes

tacaswell mentioned this pull request Apr 6, 2017

matplotlib x axis labels dissapear in scatterplot jupyter/notebook#2353

Closed

nparley added 5 commits April 9, 2017 16:13

Attempted fixes for #8113 and #10678

c81e500

Create a new method which can override _compute_plot_data if the plot…

b7036bb

… (i.e. scatter) is happy with datetimes

Move _compute_plot_data_with_dt to PlanePlot class

8b8d542

Fix lint error.

8c0ffe2

Addressed comments

13de504

Reduce line length

cbf3c73

jreback closed this Aug 1, 2017

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix Scatter plot datetime and Axis #12949

Fix Scatter plot datetime and Axis #12949

nparley commented Apr 21, 2016 •

edited by jreback

Loading

jreback commented Apr 21, 2016

sinhrks Apr 21, 2016

nparley Aug 15, 2016

jreback commented May 20, 2016

codecov-io commented May 20, 2016 •

edited by codecov bot

Loading

jorisvandenbossche commented Aug 13, 2016

sinhrks commented Aug 13, 2016

nparley commented Aug 13, 2016

nparley commented Aug 15, 2016

nparley commented Aug 16, 2016

jreback commented Sep 9, 2016

nparley commented Sep 10, 2016 •

edited

Loading

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016

jreback commented Sep 10, 2016

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016 •

edited

Loading

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016

jreback commented Nov 22, 2016

nparley commented Nov 22, 2016 •

edited

Loading

jorisvandenbossche commented Nov 22, 2016

jreback commented Jan 21, 2017

jreback commented Mar 20, 2017

jorisvandenbossche left a comment

jorisvandenbossche Mar 20, 2017

jorisvandenbossche Mar 20, 2017

jorisvandenbossche Mar 20, 2017

nparley commented Apr 9, 2017 •

edited

Loading

jreback commented May 13, 2017

jreback commented Aug 1, 2017

Fix Scatter plot datetime and Axis #12949

Fix Scatter plot datetime and Axis #12949

Conversation

nparley commented Apr 21, 2016 • edited by jreback Loading

jreback commented Apr 21, 2016

sinhrks Apr 21, 2016

Choose a reason for hiding this comment

nparley Aug 15, 2016

Choose a reason for hiding this comment

jreback commented May 20, 2016

codecov-io commented May 20, 2016 • edited by codecov bot Loading

Codecov Report

jorisvandenbossche commented Aug 13, 2016

sinhrks commented Aug 13, 2016

nparley commented Aug 13, 2016

nparley commented Aug 15, 2016

nparley commented Aug 16, 2016

jreback commented Sep 9, 2016

nparley commented Sep 10, 2016 • edited Loading

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016

jreback commented Sep 10, 2016

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016 • edited Loading

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016

jreback commented Sep 10, 2016

nparley commented Sep 10, 2016

jreback commented Nov 22, 2016

nparley commented Nov 22, 2016 • edited Loading

jorisvandenbossche commented Nov 22, 2016

jreback commented Jan 21, 2017

jreback commented Mar 20, 2017

jorisvandenbossche left a comment

Choose a reason for hiding this comment

jorisvandenbossche Mar 20, 2017

Choose a reason for hiding this comment

jorisvandenbossche Mar 20, 2017

Choose a reason for hiding this comment

jorisvandenbossche Mar 20, 2017

Choose a reason for hiding this comment

nparley commented Apr 9, 2017 • edited Loading

jreback commented May 13, 2017

jreback commented Aug 1, 2017

nparley commented Apr 21, 2016 •

edited by jreback

Loading

codecov-io commented May 20, 2016 •

edited by codecov bot

Loading

nparley commented Sep 10, 2016 •

edited

Loading

nparley commented Sep 10, 2016 •

edited

Loading

nparley commented Nov 22, 2016 •

edited

Loading

nparley commented Apr 9, 2017 •

edited

Loading