CODEWITHSUNDEEP
pythonpandasmatplotlibstatistics
Tappy
Tappy

Reputation: 87

Stacked bar plot in Pandas with max values instead of being summed

I'm trying to replicate work done in an excel pivot table. I have 3 variables (idle time, and total time, and customer). I want to make a bar plot that is stacked(overlaps) but is not a sum of all variables.

For example: if..

Total_time = 10
Idle_time = 3

I want the bar graph to show the max value of idle time and total time for each customer.

So far when I try to use stacked=True, it just stacks Total & Idle time and creates a bar graph with a max value of 10+3=13. I want two separate bars overlapping each other where one bar maxes at 10 and the other at 3.

data = df.pivot_table(df,index=['Customer'], aggfunc=np.sum)
ax = data[['Total time [hours]', 'Idle time [hours]']].plot.bar(stacked=True)
ax.set_ylabel("Miles")
ax.set_title("Total Miles Per Customer")

plt.show()

I've uploaded a screenshot of the excel version, and pandas version for further illustration.

Excel Image

Excel Image

Pandas Image

Pandas Image

Thanks in advance.

UPDATE: Ted with the solution below.

Upvotes: 1

Views: 1295

Answers (1)

Ted Petrou
Ted Petrou

Reputation: 61967

One simple solution is to plot them separately with the idle time last

df['Total time [hours]'].plot.bar(color='r')
df['Idle time [hours]'].plot.bar(color='b')

Another solution is to create a new column that is equal to the Total time minus the Idle Time and then do a stacked bar plot. You can rename the columns from there

df['Extra_Time'] = df['Total time [hours]'] - df['Idle time [hours]']
data[['Idle time [hours]', 'Extra_Time']].plot.bar(stacked=True)

Upvotes: 3

Related Questions