CODEWITHSUNDEEP
pythonpandasplotdata-science
Lee
Lee

Reputation: 791

pandas - plot bar chart for multiple categories

I have this dataframe:

Columns:

Countries
IsVertical

Example:

USA FALSE
USA FALSE
Poland FALSE
Italy TRUE
Italy TRUE

I want to create a bar chart that has 2 columns for each country (1 for True and one for false - From IsVertical) and I need the y axis to be the normalized count of times that the country appears.

I tryed something like this:

data.groupby("Country").count().plot(kind="bar")

The problem is that it doesn't give me 2 bars of IsVertical foreach country and that it only counts the number of times that IsVertical appears and doesn't divide by the total count of IsVertical

SOLUTION:

result = data.groupby('Country').apply(
         lambda group: (group.IsVertical.sum() / float(data.IsVertical.sum()))
     ).to_frame('Vertical')

    result.plot(kind='bar')

Upvotes: 0

Views: 1517

Answers (1)

Alexander
Alexander

Reputation: 109546

result = df.groupby('Countries').apply(
             lambda group: (group.IsVertical == 'TRUE').sum() / 
                            float(group.IsVertical.count())
         ).to_frame('Vertical')

result['NotVertical'] = 1 - result.Vertical
>>> result
           Vertical  NotVertical
Countries                       
Italy             1            0
Poland            0            1
USA               0            1

result.plot(kind='bar')

Upvotes: 1

Related Questions