Series.plot should label y axis with Series name

Code Sample, a copy-pastable example if possible

import pandas as pd
import matplotlib.pyplot as plt
ind = pd.Series(['first', 'second', 'third'], name='count')
s = pd.Series([1, 2, 3], index=ind, name='i should be a y label')
s.plot()
plt.show()

Problem description

So when you call Series.plot, xlabel is set to the name of the index, but ylabel is not set to the name of the series itself. To me it seems asymmetric and believe it should set ylabel automatically. Also, if I plot it with s.plot(kind='hist'), I get ylabel “Frequency” but no xlabel, which I would expect to be the name of the series as well.

Expected Output

ylabel set to name of the series for usual plot, and xlabel set to name of the series for hist plot

Output of pd.show_versions()

[paste the output of pd.show_versions() here below this line] INSTALLED VERSIONS

commit: None python: 3.6.2.final.0 python-bits: 64 OS: Windows OS-release: 10 machine: AMD64 processor: AMD64 Family 16 Model 4 Stepping 3, AuthenticAMD byteorder: little LC_ALL: None LANG: None LOCALE: None.None

pandas: 0.20.3 pytest: 3.2.1 pip: 9.0.1 setuptools: 36.5.0.post20170921 Cython: 0.26.1 numpy: 1.13.1 scipy: 0.19.1 xarray: None IPython: 6.1.0 sphinx: 1.6.3 patsy: 0.4.1 dateutil: 2.6.1 pytz: 2017.2 blosc: None bottleneck: 1.2.1 tables: 3.4.2 numexpr: 2.6.2 feather: None matplotlib: 2.0.2 openpyxl: 2.4.8 xlrd: 1.1.0 xlwt: 1.3.0 xlsxwriter: 0.9.8 lxml: 3.8.0 bs4: 4.6.0 html5lib: 0.999999999 sqlalchemy: 1.1.13 pymysql: None psycopg2: None jinja2: 2.9.6 s3fs: None pandas_gbq: None pandas_datareader: None

[Eoksni]

Actually, I made my workaround even more convenient. I just have this snippet at the start of my python notebook, and then I can use Series.plot in ordinal way, but it now puts ylabel as I wanted.

from functools import wraps

def monkey_patch_series_plot():
    """Makes Series.plot to show series name as ylabel/xlabel according to plot kind."""
    f = pd.Series.plot.__call__
    @wraps(f)
    def _decorator(*kargs, **kwargs):
        res = f(*kargs, **kwargs)
        s = kargs[0].__dict__['_data']
        if s.name:
            try:
                kind = kwargs['kind']
            except KeyError:
                kind = 'line'
            if kind == 'line' or kind == 'scatter':
                plt.ylabel(s.name)
            elif kind == 'hist':
                plt.xlabel(s.name)
        return res
    pd.Series.plot.__call__ = _decorator
monkey_patch_series_plot()

So, I’m entirely satisfied with this so I close the issue.

[Eoksni]

@yudai-nkt I don’t think you are doing something wrong, your exact code piece worked flawlessly for me back in the days. I haven’t been using pandas for a while and not exactly sure how it can be fixed for newer version.

As a suggestion, you can put a breakpoint before kargs[0].__dict__ line and see what is inside and change the code accordingly. And yeah, all of this is fiddling with internal data structures and such structure probably can be changed even in minor releases as it is not part of public API.