I'm trying to create a stacked bar chart, with
xaxis = 'customer_id(count), yaxis = 'age_band',
and the 3 different loyalty groups stacked in the chart (hue), so I should see 6 bars each with 2-3 different colours.
code I've tried:
df.groupby(['age_band','loyalty']).agg(Loyalty=('customer_id', pd.Series.nunique)).plot(kind='bar', stacked=True)
Picture for reference:

CodePudding user response:
IIUC use DataFrame.pivot_table:
(df.pivot_table(index='age_band',columns='loyalty', values='customer_id', aggfunc='nunique')
.plot(kind='bar', stacked=True))
