pandas dataframe hexbin plot has no xlabel or axis values

Question

I am trying to plot a dataframe as hexbin but can't seem to get the xlabel and x-axis values to plot. The dataframe is as follows:

szen_df.xs('left', level='pos')

enter image description here

and the plot is as follows:

szen_df.xs('left', level='pos').plot(x='szen', y='lat', xlim=(0,90), ylim=(-90,90), kind='hexbin', colormap='Reds' )

with result:

enter image description here

This works find by me. Perhaps you should post a minimal DataFrame that exhibits the problem. — Ami Tavory, Jul 04 '15 at 12:52
Works for me too. Let's have your `pandas` version and your `python` version. See if that gets us anywhere. — LondonRob, Jul 04 '15 at 13:16
I tried it on a minimal simulated dataframe and it works. Just seems to be the larger dataframe. Would having nan in the dataset have anything to do with this? — Jacques MALAPRADE, Jul 04 '15 at 16:34
It seems to work if I plot using matplotlib directly instead of through pandas `df.plot()`, but only if I use `df.dropna()`. I suspect that pandas won't work with nan in the dataset. I tried `df.dropna().xs(blabla).plot(blabla)` but won't accept it. I still have the problem that I can't just select the `df..xs('left', level='pos')` values, so still looking for a solution. — Jacques MALAPRADE, Jul 04 '15 at 16:53
Is there a way to use `dropna()` and something like `xs()` together? — Jacques MALAPRADE, Jul 04 '15 at 17:01
What's the error you're getting? One possible solution would be new_df = df.dropna(). then your plot statement with new_df — Bob Haffner, Jul 04 '15 at 17:10
I tried the `new_df = df.dropna()` and still doesn't plot the x-axis label and values. No error is printed. — Jacques MALAPRADE, Jul 05 '15 at 04:40
Are you using IPython Notebook? I think it may cause the issue for me. — Lev Levitsky, Jul 23 '15 at 14:46
I submitted a [bug report](https://github.com/ipython/ipython/issues/8653), let's see how it goes. — Lev Levitsky, Jul 23 '15 at 15:32

score 30 · Answer 1 · answered Jul 26 '15 at 03:34

30

I'm seeing this problem as well with python 2.7, pandas 0.16.2.dev, and ipython version 3.1.0.

Looking to the documentation for pandas.DataFrame.plot, the only reference to the x labels and ticks being invisible is in the sharex option. So, I manually set sharex=False, and poof, the xlabel and xticks come back.

Clearly this is a bug, but I hope this helps for the time being. I'll add these comments to the bug report that you've already started.

answered Jul 26 '15 at 03:34

Bryan Dannowitz

1,186
6
6

4

Still present in 0.18.1, relevant [Issue on github](https://github.com/pandas-dev/pandas/issues/10678) – nepix32 Nov 16 '16 at 16:42
STILL present in 1.1.5. https://github.com/pandas-dev/pandas/issues/36064 – Jesuisme Apr 06 '21 at 11:41
STILL present in 1.4.1 – Olivier D'Ancona Mar 25 '22 at 12:00

score 1 · Answer 2 · edited May 23 '17 at 10:31

1

Looks like the same as the issue I had here: matplotlib scatterplot x axis labels

The workaround which worked was to set the axes explicitly. Just leaving this here for the future :)

Might be connected with https://github.com/pandas-dev/pandas/pull/12949 , Open as of now (6th of April 17)

edited May 23 '17 at 10:31

Community

1
1

answered Apr 06 '17 at 18:33

CatInABox

387
2
4
12

pandas dataframe hexbin plot has no xlabel or axis values

2 Answers2

Linked