scatter plot and hexbin plot lose x-axis when colorbar is included. #20446

javadnoorb · 2018-03-22T02:47:27Z

The x-axis for scatter plot and hexbin plot disappears when colorbar is included. This seems to be because colorbar axis is looped through in _handle_shared_axes:

pandas/pandas/plotting/_core.py

Line 426 in 8a58303

_handle_shared_axes(axarr=all_axes, nplots=len(all_axes),

after discussing with @TomAugspurger (issue #10611), we decided to try adding the attribute __is_pandas_colorbar to colorbar axis object and skipping it during handling of shared axes.

I've done some tests that seem to fix the issue. But we may need more tests:

%matplotlib inline
import matplotlib.pylab as pl
from mypandas import pandas as pd
import numpy as np
random_array = np.random.random((1000,3))

df = pd.DataFrame(random_array,columns=['A label','B label','C label'])
df.plot.scatter('A label','B label',c='C label', PATCH_MODE_FLAG = False);pl.title('pandas current version');
df.plot.scatter('A label','B label',c='C label', PATCH_MODE_FLAG = True);pl.title('patch fixing x-axis');

df.plot.hexbin('A label','B label', gridsize=25, PATCH_MODE_FLAG = False);pl.title('pandas current version');
df.plot.hexbin('A label','B label', gridsize=25, PATCH_MODE_FLAG = True);pl.title('patch fixing x-axis');

… and hexbin

TomAugspurger · 2018-03-22T11:09:08Z

Thanks, let us know when you're ready for us to take a look.

How does this work with subplots=True? Do we not have to worry about that since it's a scatter plot?

How does this work with user-specified axes via ax=? Especially with multiple subplots. It may already be broken, so no worries if it doesn't work well.

javadnoorb · 2018-03-22T13:31:16Z

Yes, it's already broken. Please feel free to take a look:

def make_multiplots(numrows=2,numcols=2,sharey=True,sharex=True,PATCH_MODE_FLAG = True, is_scatter = True):
    fig, axes = pl.subplots(numrows,numcols,sharey=sharey,sharex=sharex )
    for i in range(axes.shape[0]):
        for j in range(axes.shape[1]):
            random_array = np.random.random((1000,3))
            df = pd.DataFrame(random_array,columns=['A label','B label','C label'])
            if is_scatter:
                df.plot.scatter('A label','B label',c='C label', PATCH_MODE_FLAG = PATCH_MODE_FLAG,ax=axes[i,j]);
            else:
                df.plot.hexbin('A label','B label',gridsize=20, PATCH_MODE_FLAG = PATCH_MODE_FLAG,ax=axes[i,j]);
            
            
make_multiplots(numrows=2,numcols=2,sharey=True, sharex=True, PATCH_MODE_FLAG = False, is_scatter=True);pl.suptitle('pandas current version');            
make_multiplots(numrows=2,numcols=2,sharey=True, sharex=True, PATCH_MODE_FLAG = True, is_scatter=True);pl.suptitle('patch fixing x-axis');            
make_multiplots(numrows=2,numcols=2,sharey=True, sharex=True, PATCH_MODE_FLAG = False, is_scatter=False);pl.suptitle('pandas current version');            
make_multiplots(numrows=2,numcols=2,sharey=True, sharex=True, PATCH_MODE_FLAG = True, is_scatter=False);pl.suptitle('patch fixing x-axis');

TomAugspurger

I would say make a new issue for subplots, and ignore it for now.

As for testing, how about we make two plots, one with c and one without, and assert that the tick labels are the same?

fig, ax = plt.supblots()
df.plot.scatter(x='x', y='y', ax=ax)

fig, ax2 = plt.subplots()
df.plot.scatter(x='x', y='y', c='c', ax=ax2)

assert ax.get_xticklabels() == ax2.get_xticklabels()

or something like that.

TomAugspurger · 2018-03-22T13:37:17Z

pandas/plotting/_core.py

@@ -870,6 +870,8 @@ def _make_plot(self):
        scatter = ax.scatter(data[x].values, data[y].values, c=c_values,
                             label=label, cmap=cmap, **self.kwds)
        if cb:
+            if PATCH_MODE:


I assume this ``PATCH_MODE` is just for debugging?

yes, it should be removed. Didn't know of a better way to do this.

TomAugspurger · 2018-03-22T13:38:01Z

pandas/plotting/_tools.py

@@ -298,20 +298,20 @@ def _remove_labels_from_axis(axis):

 def _handle_shared_axes(axarr, nplots, naxes, nrows, ncols, sharex, sharey):
    if nplots > 1:
-
+        


This whitespace will likely fail the linter

TomAugspurger · 2018-03-22T13:38:08Z

pandas/plotting/_tools.py

        if nrows > 1:
            try:
                # first find out the ax layout,
                # so that we can correctly handle 'gaps"
                layout = np.zeros((nrows + 1, ncols + 1), dtype=np.bool)
                for ax in axarr:
                    layout[ax.rowNum, ax.colNum] = ax.get_visible()
-
+                    


javadnoorb · 2018-03-22T13:52:10Z

As of subplots=True, I've only used it with histograms and line plots. I'm not familiar with their usage with scatter plots/hexbin plots. So I'm not sure how to test them. Simply using it with single plot doesn't interfere with the outcome, but I'm assuming that's not what you were referring to. I think we'll need your expertise here.

df.plot.scatter('A label','B label',c='C label', PATCH_MODE_FLAG = True, subplots=True);

pep8speaks · 2018-03-22T14:01:51Z

Hello @javadnoorb! Thanks for updating the PR.

Cheers ! There are no PEP8 issues in this Pull Request. 🍻

Comment last updated on July 04, 2018 at 03:04 Hours UTC

codecov · 2018-03-22T14:02:04Z

Codecov Report

Merging #20446 into master will increase coverage by 0.02%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #20446      +/-   ##
==========================================
+ Coverage    91.9%   91.92%   +0.02%     
==========================================
  Files         154      158       +4     
  Lines       49657    49705      +48     
==========================================
+ Hits        45638    45693      +55     
+ Misses       4019     4012       -7

Flag	Coverage Δ
#multiple	`90.3% <100%> (+0.02%)`	⬆️
#single	`41.99% <11.11%> (+0.09%)`	⬆️

Impacted Files	Coverage Δ
pandas/plotting/_core.py	`83.57% <100%> (+0.03%)`	⬆️
pandas/core/indexes/datetimes.py	`95.48% <0%> (-0.18%)`	⬇️
pandas/core/indexes/timedeltas.py	`91.08% <0%> (-0.16%)`	⬇️
pandas/core/indexes/period.py	`92.53% <0%> (-0.14%)`	⬇️
pandas/core/dtypes/cast.py	`88.36% <0%> (-0.13%)`	⬇️
pandas/core/indexes/accessors.py	`90.09% <0%> (-0.1%)`	⬇️
pandas/core/internals.py	`95.53% <0%> (-0.07%)`	⬇️
pandas/core/arrays/__init__.py	`100% <0%> (ø)`	⬆️
pandas/core/arrays/timedelta.py	`100% <0%> (ø)`
pandas/core/arrays/period.py	`100% <0%> (ø)`
... and 17 more

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 dc45fba...7196d6e. Read the comment docs.

javadnoorb

removed the whitespaces

TomAugspurger · 2018-03-22T14:03:42Z

Don't worry about subplots. It would be something like df.plot.scatter(x=0, y=[1, 2], subplots=True), which currently isn't allowed.

javadnoorb · 2018-03-22T18:02:46Z

That test with assert doesn't deal with the issue. It seems like if you make a subplot and pass the axis to pandas, the x-axis comes out fine. I didn't use environments, and don't know how to switch from my development pandas to the native version without an uninstall. But with the PATCH_MODE_FLAG=False you can see it:

fig, ax1 = pl.subplots()
df.plot.scatter(x='A label', y='B label', c='C label',ax=ax1, PATCH_MODE_FLAG=False)

df.plot.scatter(x='A label', y='B label', c='C label', PATCH_MODE_FLAG=False)

javadnoorb · 2018-03-22T18:18:49Z

This works:

PATCH_MODE_FLAG = False
ax1 = df.plot.scatter(x='A label', y='B label', PATCH_MODE_FLAG=PATCH_MODE_FLAG)
ax2 = df.plot.scatter(x='A label', y='B label', c='C label', PATCH_MODE_FLAG=PATCH_MODE_FLAG)
assert (ax1.get_xticks()==ax2.get_xticks()).all(), "axis labels are not the same"

but regardless of PATCH_MODE_FLAG value it doesn't throw an error.

I think the issue is not with the xticks attribute but rather showing them.

javadnoorb · 2018-03-22T18:26:28Z

Trying a few of these. Quite bizarre:

PATCH_MODE_FLAG = False
ax1 = df.plot.scatter(x='A label', y='B label', PATCH_MODE_FLAG=PATCH_MODE_FLAG)
ax2 = df.plot.scatter(x='A label', y='B label', c='C label', PATCH_MODE_FLAG=PATCH_MODE_FLAG)
assert (ax1.get_xticks()==ax2.get_xticks()).all(), "Axis xticks are not the same"
assert ax1.xaxis.get_visible()==ax2.xaxis.get_visible(), "x-axis visbilities are not the same"
print(ax1.get_xticklabels(),ax2.get_xticklabels())
print(ax1.get_xticks(),ax2.get_xticks())
print(ax1.xaxis.get_visible(),ax2.xaxis.get_visible())

TomAugspurger · 2018-03-28T16:27:38Z

@javadnoorb how's this going? For this PR I would ignore user-provided axes, and only worry about the common case of df.plot(x='x', y='y', c='c').

javadnoorb · 2018-03-28T16:48:06Z

@TomAugspurger I haven't worked on this for a few days. Let me see if I can come up with a reasonable test. get_xticklabels returns an object, so I think we can't use it for comparison. I think the reason the x-labels disappear should be because of these three lines:

pandas/pandas/plotting/_tools.py

Line 282 in 8a58303

t.set_visible(False)

pandas/pandas/plotting/_tools.py

Line 296 in 8a58303

axis.get_label().set_visible(False)

pandas/pandas/plotting/_tools.py

Line 293 in 8a58303

t.set_visible(False)

my guess is that something like:

assert ax1.xaxis.get_majorticklabels().get_visible()==ax2.xaxis.get_majorticklabels().get_visible()

and

assert ax1.xaxis.get_label().get_visible()==ax2.xaxis.get_label().get_visible()

and

assert ax1.xaxis.get_minorticklabels().get_visible()==ax2.xaxis.get_minorticklabels().get_visible()

or something like that, should work for the test.

Let me give it a shot.

TomAugspurger · 2018-03-28T16:57:27Z

Sounds good. Let me know if that doesn't work and I can take a look.

…

On Wed, Mar 28, 2018 at 11:48 AM, Javad Noorbakhsh ***@***.*** > wrote: @TomAugspurger <https://github.com/TomAugspurger> I haven't worked on this for a few days. Let me see if I can come up with a reasonable test. get_xticklabels returns an object, so I think we can't use it for comparison. I think the reason the x-labels disappear should be because of these two lines: https://github.com/pandas-dev/pandas/blob/8a5830305d2d9a087fbe46bd968218 104ffdfc49/pandas/plotting/_tools.py#L282 https://github.com/pandas-dev/pandas/blob/8a5830305d2d9a087fbe46bd968218 104ffdfc49/pandas/plotting/_tools.py#L296 my guess is that something like: assert ax1.get_majorticklabels().get_visible()==ax2.get_ majorticklabels().get_visible() and assert ax1.get_label().get_visible()==ax2.get_label().get_visible() or something like that, should work for the test. Let me give it a shot. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#20446 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABQHIkLc6e7wel8eSgSbmlptCfR_HWFaks5ti77NgaJpZM4S2XhP> .

javadnoorb · 2018-03-28T18:07:17Z

Ok, I wrote a test that seems to work:

def test_if_colorbar_affects_xaxis_visibility(PATCH_MODE_FLAG=True):
    random_array = np.random.random((1000,3))
    df = pd.DataFrame(random_array,columns=['A label','B label','C label'])

    ax1 = df.plot.scatter(x='A label', y='B label', PATCH_MODE_FLAG=PATCH_MODE_FLAG)
    ax2 = df.plot.scatter(x='A label', y='B label', c='C label', PATCH_MODE_FLAG=PATCH_MODE_FLAG)

    assert all([vis[0].get_visible()==vis[1].get_visible() for vis in zip(ax1.xaxis.get_minorticklabels(),ax2.xaxis.get_minorticklabels())]), 'minor x-axis tick labels visibility changes when colorbar included'
    assert all([vis[0].get_visible()==vis[1].get_visible() for vis in zip(ax1.xaxis.get_majorticklabels(),ax2.xaxis.get_majorticklabels())]), 'major x-axis tick labels visibility changes when colorbar included'
    assert ax1.xaxis.get_label().get_visible()==ax2.xaxis.get_label().get_visible(),'x-axis label visibility changes when colorbar included'

test_if_colorbar_affects_xaxis_visibility(PATCH_MODE_FLAG=True)

test_if_colorbar_affects_xaxis_visibility(PATCH_MODE_FLAG=False)

AssertionError                            Traceback (most recent call last)
<ipython-input-47-0cc25b4e00f4> in <module>()
     10     assert ax1.xaxis.get_label().get_visible()==ax2.xaxis.get_label().get_visible(),'x-axis label visibility changes when colorbar included'
     11 
---> 12 test_if_colorbar_affects_xaxis_visibility(PATCH_MODE_FLAG=False)

<ipython-input-47-0cc25b4e00f4> in test_if_colorbar_affects_xaxis_visibility(PATCH_MODE_FLAG)
      7 
      8     assert all([vis[0].get_visible()==vis[1].get_visible() for vis in zip(ax1.xaxis.get_minorticklabels(),ax2.xaxis.get_minorticklabels())]), 'minor x-axis tick labels visibility changes when colorbar included'
----> 9     assert all([vis[0].get_visible()==vis[1].get_visible() for vis in zip(ax1.xaxis.get_majorticklabels(),ax2.xaxis.get_majorticklabels())]), 'major x-axis tick labels visibility changes when colorbar included'
     10     assert ax1.xaxis.get_label().get_visible()==ax2.xaxis.get_label().get_visible(),'x-axis label visibility changes when colorbar included'
     11 

AssertionError: major x-axis tick labels visibility changes when colorbar included

If that seems fine I'll include the changes and remove the PATCH_MODE flag.

I'm not familiar with writing tests. Is this where it goes?
https://github.com/pandas-dev/pandas/blob/master/pandas/tests/plotting/test_misc.py

TomAugspurger · 2018-03-28T18:39:10Z

That test seems sensible.

Tests can go in pandas/pandas/tests/plotting/test_frame.py under TestFrame. There are some other ones there for scatter.

removing PATH_MODE debugging flag

javadnoorb · 2018-03-28T18:54:06Z

Ok, great. I'll check if it passes the regular checks before including the test.

remove whitespaces

javadnoorb · 2018-03-28T19:23:08Z

Also need a test for hexbin. This one seems to work:

def test_if_hexbin_xaxis_label_is_visible(PATCH_MODE_FLAG=True):
    random_array = np.random.random((1000,3))
    df = pd.DataFrame(random_array,columns=['A label','B label','C label'])

    ax = df.plot.hexbin('A label','B label', gridsize=12, PATCH_MODE_FLAG = PATCH_MODE_FLAG);    
    assert all([vis.get_visible() for vis in ax.xaxis.get_minorticklabels()]), 'minor x-axis tick labels are not visible'
    assert all([vis.get_visible() for vis in ax.xaxis.get_majorticklabels()]), 'major x-axis tick labels are not visible'
    assert ax.xaxis.get_label().get_visible(),'x-axis label is not visible'

test_if_hexbin_xaxis_label_is_visible(PATCH_MODE_FLAG=True)

test_if_hexbin_xaxis_label_is_visible(PATCH_MODE_FLAG=False)

---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
<ipython-input-66-cd06ece40a4d> in <module>()
      9 
     10 
---> 11 test_if_hexbin_xaxis_label_is_visible(PATCH_MODE_FLAG=False)

<ipython-input-66-cd06ece40a4d> in test_if_hexbin_xaxis_label_is_visible(PATCH_MODE_FLAG)
      4 
      5     ax = df.plot.hexbin('A label','B label', gridsize=12, PATCH_MODE_FLAG = PATCH_MODE_FLAG);
----> 6     assert all([vis.get_visible() for vis in ax.xaxis.get_minorticklabels()]), 'minor x-axis tick labels are not visible'
      7     assert all([vis.get_visible() for vis in ax.xaxis.get_majorticklabels()]), 'major x-axis tick labels are not visible'
      8     assert ax.xaxis.get_label().get_visible(),'x-axis label is not visible'

AssertionError: minor x-axis tick labels are not visible

javadnoorb · 2018-03-28T23:02:35Z

These changes seem pretty benign. Is there any specific reason for Travis-CI to throw an error?

TomAugspurger · 2018-03-29T00:06:49Z

pandas/plotting/_tools.py

@@ -311,7 +311,7 @@ def _handle_shared_axes(axarr, nplots, naxes, nrows, ncols, sharex, sharey):
                    # only the last row of subplots should get x labels -> all
                    # other off layout handles the case that the subplot is
                    # the last in the column, because below is no subplot/gap.
-                    if not layout[ax.rowNum + 1, ax.colNum]:
+                    if not layout[ax.rowNum + 1, ax.colNum] or getattr(ax, '_pandas_colorbar_axes', False):


This line is too long: https://travis-ci.org/pandas-dev/pandas/jobs/359516939#L3006

Thanks. I'll fix it. Sorry I'm pretty new to the open-source world.

…s not exclude x-axis label and tick-labels

javadnoorb · 2018-04-10T11:07:52Z

@TomAugspurger. Seems like the issue is because in the inline backend floats corresponding to location of axes have too many arbitrary trailing digits. For example two axes that have their bottom left corners at 0.3 may appear as 0.3 and 0.30000001. I rounded them in _core.py.

This new commit works for single plots with pdf and inline backends. It still fails for multiplots. We'll need to dig deeper for that. But it seems like we don't need the previous hack anymore.

TomAugspurger · 2018-04-12T19:24:40Z

pandas/plotting/_core.py

+            #  `get_points()` have too many ad hoc trailing
+            # digits. Unless rounded these values won't
+            # match in the following set operations
+            points = np.round(points, 3)


I worry about this causing debugging nightmares further down the road... I'd prefer to solve this local to the problem if possible.

If not, you could perhaps include a keyword to _get_axes_layout? I'm not sure how difficult that would be to plumb through.

I agree. That's pretty tough to track. Might be better to avoid rounding off. Especially that 3 digit rounding is pretty arbitrary and may break in unknown ways.

I made a new commit. The changes are local now. I'm forcing colorbars to take on their relevant axes positions from the parent axes. The merge passes the tests and resolves the xlabel issue (#10611 and #10678) but not with the subplots (#20455).

* minor change in comment * added condition on inline backend * added condition on inline backend * issue with floats determining the position of axes corners in ipython * fixing axis position roundoff error locally

* minor change in comment * added condition on inline backend * added condition on inline backend * issue with floats determining the position of axes corners in ipython * fixing axis position roundoff error locally * broken merge

javadnoorb · 2018-06-27T21:49:43Z

I just remembered about this. The last commit seemed to have passed all the checks. Is there any interest to follow up on it?

TomAugspurger

Sorry for dropping this. Can you merge in master and move the whatsnew to 0.24.0?

Looks good otherwise, so ping on green.

TomAugspurger · 2018-06-29T21:35:52Z

doc/source/whatsnew/v0.23.0.txt

@@ -1176,7 +1176,7 @@ Plotting
 - Bug in formatting tick labels with ``datetime.time()`` and fractional seconds (:issue:`18478`).
 - :meth:`Series.plot.kde` has exposed the args ``ind`` and ``bw_method`` in the docstring (:issue:`18461`). The argument ``ind`` may now also be an integer (number of sample points).
 - :func:`DataFrame.plot` now supports multiple columns to the ``y`` argument (:issue:`19699`)
-
+- Bug in :func:'DataFrame.plot.scatter' and :func:'DataFrame.plot.hexbin' caused x-axis label and ticklabels to disappear when colorbar was on in IPython inline backend (:issue:`10611` and :issue:`10678`)


This should be moved to 0.24.0

TomAugspurger · 2018-06-29T21:36:41Z

pandas/plotting/_core.py

@@ -882,8 +882,13 @@ def _make_plot(self):
            kws = dict(ax=ax)
            if self.mpl_ge_1_3_1():
                kws['label'] = c if c_is_column else ''
-            self.fig.colorbar(img, **kws)
-
+            cbar = self.fig.colorbar(img, **kws)


Maybe refactor this and https://github.com/pandas-dev/pandas/pull/20446/files#diff-2b118bda866f3d626ceb6b529c62cd1aR932 to a method. They're identical, right? I guess aside from **kws vs. ax=ax?

Merge branch 'master' of https://github.com/pandas-dev/pandas

javadnoorb · 2018-06-30T22:20:46Z

Thanks @TomAugspurger. I cleaned up the code. It's passing the tests. Here's what the output in jupyterlab looks like:

%matplotlib inline
import numpy as np
import pandas as pd
random_array = np.random.random((1000, 3))
df = pd.DataFrame(random_array,columns=['A label','B label','C label'])
df.plot.scatter('A label', 'B label', c='C label')

df.plot.hexbin('A label','B label', gridsize=12);

jreback

comments, otherwise lgtm. ping on green.

jreback · 2018-07-04T00:30:23Z

pandas/plotting/_core.py

@@ -833,6 +833,23 @@ def _post_plot_logic(self, ax, data):
        ax.set_ylabel(pprint_thing(y))
        ax.set_xlabel(pprint_thing(x))

+    def _plot_colorbar(self, kws):


can you make ax a named arg here and so you would pass as **kwds when you call.

jreback · 2018-07-04T00:31:31Z

pandas/plotting/_core.py

+        # To deal with this, this method forces the colorbar
+        # height to take the height of the parent axes.
+        ax = kws['ax']
+        img = ax.collections[0]


is there an issue we can reference about this impl. IOW next person to look at wouldn't know where these set_position come from. so a) ref this issue in a comment and b) an Ipython refernce (if you can)

jreback · 2018-07-04T12:59:46Z

thanks @javadnoorb very nice!

javadnoorb · 2018-07-04T13:01:54Z

thanks @jreback!

…andas-dev#20446)

javadnoorb added 3 commits March 21, 2018 20:53

removed colorbars from _handle_shared_axes when called by scatterplot…

e2abff6

… and hexbin

removed colorbars from _handle_shared_axes when called by scatterplot…

0ea226f

… and hexbin

added a debug global variable

80f949f

TomAugspurger added the Visualization plotting label Mar 22, 2018

TomAugspurger reviewed Mar 22, 2018

View reviewed changes

removed whitespaces

8f52eab

javadnoorb commented Mar 22, 2018

View reviewed changes

javadnoorb mentioned this pull request Mar 22, 2018

xlabel disappears in scatter plots with subplots and colorbar #20455

Closed

Update _core.py

215fb89

removing PATH_MODE debugging flag

Update _core.py

7c6efe2

remove whitespaces

TomAugspurger reviewed Mar 29, 2018

View reviewed changes

javadnoorb and others added 2 commits March 28, 2018 22:52

Update _tools.py

8c71f60

included tests for scatterplot and hexbin plot to ensure colorbar doe…

63b9ee6

…s not exclude x-axis label and tick-labels

javadnoorb added 4 commits April 10, 2018 06:51

Update _core.py

1fb4eec

Update _core.py

0397999

Update v0.23.0.txt

5ba105f

Update _core.py

ad1b495

TomAugspurger reviewed Apr 12, 2018

View reviewed changes

javadnoorb and others added 4 commits April 12, 2018 23:23

Inline backend - fixing colorbar axis position roundoff error locally

bdc7f5a

* minor change in comment * added condition on inline backend * added condition on inline backend * issue with floats determining the position of axes corners in ipython * fixing axis position roundoff error locally

broken merge

e656f6e

* minor change in comment * added condition on inline backend * added condition on inline backend * issue with floats determining the position of axes corners in ipython * fixing axis position roundoff error locally * broken merge

Merge remote-tracking branch 'upstream/master'

1e4efd3

Merge remote-tracking branch 'upstream/master'

45aea6f

TomAugspurger reviewed Jun 29, 2018

View reviewed changes

javadnoorb added 7 commits June 30, 2018 11:57

update fork

eeb6a9f

Merge branch 'master' of https://github.com/pandas-dev/pandas

refactoring

a802ecf

refactoring

8d766ae

updating whatsnew docs

4f380b2

updating whatsnew docs

04ff768

updating whatsnew docs

8c0f2e5

updating whatsnew docs

564790f

jreback added this to the 0.24.0 milestone Jul 4, 2018

jreback requested changes Jul 4, 2018

View reviewed changes

javadnoorb mentioned this pull request Jul 4, 2018

The height of axes of colorbars and parent plots do not exactly match in the inline backend ipython/ipython#11215

Open

added more comments

7196d6e

jreback approved these changes Jul 4, 2018

View reviewed changes

jreback merged commit 62a0ebc into pandas-dev:master Jul 4, 2018

Sup3rGeo pushed a commit to Sup3rGeo/pandas that referenced this pull request Oct 1, 2018

scatter plot and hexbin plot lose x-axis when colorbar is included. (p…

a3406e9

…andas-dev#20446)

stelios-c mentioned this pull request Oct 16, 2021

BUG: Hexbin plots does not display x label and xtick labels in pandas 1.3.3 #44050

Closed

3 tasks

		@@ -298,20 +298,20 @@ def _remove_labels_from_axis(axis):

		def _handle_shared_axes(axarr, nplots, naxes, nrows, ncols, sharex, sharey):
		if nplots > 1:

scatter plot and hexbin plot lose x-axis when colorbar is included. #20446

scatter plot and hexbin plot lose x-axis when colorbar is included. #20446

Conversation

javadnoorb commented Mar 22, 2018 • edited by jreback Loading

TomAugspurger commented Mar 22, 2018

javadnoorb commented Mar 22, 2018 • edited Loading

TomAugspurger left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

javadnoorb commented Mar 22, 2018

pep8speaks commented Mar 22, 2018 • edited Loading

Comment last updated on July 04, 2018 at 03:04 Hours UTC

codecov bot commented Mar 22, 2018 • edited Loading

Codecov Report

javadnoorb left a comment

Choose a reason for hiding this comment

TomAugspurger commented Mar 22, 2018

javadnoorb commented Mar 22, 2018

javadnoorb commented Mar 22, 2018

javadnoorb commented Mar 22, 2018

TomAugspurger commented Mar 28, 2018

javadnoorb commented Mar 28, 2018 • edited Loading

TomAugspurger commented Mar 28, 2018 via email

javadnoorb commented Mar 28, 2018

TomAugspurger commented Mar 28, 2018

javadnoorb commented Mar 28, 2018

javadnoorb commented Mar 28, 2018

javadnoorb commented Mar 28, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

javadnoorb commented Apr 10, 2018

Choose a reason for hiding this comment

javadnoorb Apr 13, 2018 • edited Loading

Choose a reason for hiding this comment

javadnoorb commented Jun 27, 2018

TomAugspurger left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

javadnoorb commented Jun 30, 2018 • edited Loading

jreback left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback commented Jul 4, 2018

javadnoorb commented Jul 4, 2018

javadnoorb commented Mar 22, 2018 •

edited by jreback

Loading

javadnoorb commented Mar 22, 2018 •

edited

Loading

pep8speaks commented Mar 22, 2018 •

edited

Loading

codecov bot commented Mar 22, 2018 •

edited

Loading

javadnoorb commented Mar 28, 2018 •

edited

Loading

javadnoorb Apr 13, 2018 •

edited

Loading

javadnoorb commented Jun 30, 2018 •

edited

Loading