CODEWITHSUNDEEP
pythonpandasseaborn
Neggor
Neggor

Reputation: 151

Is there any way to plot several barplots by groupby?

train['Name_title'].groupby(pd.qcut(train['Name_Len'],5)).value_counts(normalize = True)

Part of the output is:

Name_Len        Name_title
(11.999, 19.0]  Mr.           0.803922
                Miss.         0.151961
                Mrs.          0.019608
                Master.       0.009804
                Col.          0.004902
                Dr.           0.004902
                Rev.          0.004902
(19.0, 23.0]    Mr.           0.698718
                Miss.         0.237179
                Master.       0.025641
                Mrs.          0.019231
                Dr.           0.006410
                Mlle.         0.006410
                Rev.          0.006410

And I want to plot the proportion of each title for each group. Is there any way to do it straightforward?

Thank you in advance.

Upvotes: 0

Views: 40

Answers (1)

StupidWolf
StupidWolf

Reputation: 46898

You can use cross tab to tabulate and use the pandas plot with stacked=True :

np.random.seed(111)

train = pd.DataFrame({'Name_title':np.random.choice(['Mr','Mrs','Miss'],20),
                     'Name_Len':np.random.uniform(1,50,20)})

pd.crosstab(pd.qcut(train['Name_Len'],5),
            train['Name_title'],normalize='index').plot.bar(stacked=True)

enter image description here

If you unstack it will work too:

train['Name_title'].groupby(
    pd.qcut(train['Name_Len'],5)
).value_counts(normalize = True).unstack().plot.bar(stacked=True)

Upvotes: 2

Related Questions