How to create graphs of relative frequency from pandas dataframe

Question

I know that it's possible to create a histogram from a pandas dataframe column with matplotlib.pyplot using the code:

df.plot.hist(y='Distance')

Which creates a graph like this:

However what I'm looking for is a plot of relative frequency, expressed as a percentage of the total. I'd also like for the graph to have an overflow bin at 300 so that it looks something along the lines of:

cleuton · Accepted Answer · 2017-10-16 11:55:16Z

3

Try this:

orders = [{'number': 1029,'brand':'XPTO','qty':50},
      {'number': 3233,'brand':'ABCD','qty':50},
      {'number': 5455,'brand':'XPTO','qty':50},
      {'number': 1234,'brand':'ABCD','qty':50},
      {'number': 7654,'brand':'TXWZ','qty':50},
      {'number': 8765,'brand':'XPTO','qty':50},
      {'number': 4354,'brand':'TXWZ','qty':50},
      {'number': 9089,'brand':'XPTO','qty':50},
      {'number': 1031,'brand':'XPTO','qty':50}]
orders_df = pd.DataFrame(orders)
series = orders_df['brand'].value_counts() / len(orders_df)
indx = [0,1,2]
plt.bar(indx, series*100)
plt.ylabel('%')
plt.title('Relative fequency')
plt.xticks(indx, series.index)

answered Oct 16, 2017 at 11:55

cleuton

1651 silver badge5 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

BlueCoder Over a year ago

You can use series = orders_df['brand'].value_counts(True) which normalizes automatically without having to divide by len(orders_df)

user299791 Over a year ago

since you *100, these are now percentage frequencies

Collectives™ on Stack Overflow

How to create graphs of relative frequency from pandas dataframe

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related