Matplotlib sorting using index as bar denominator

Question

I am at an impasse and cannot find a solution. I wish to generate a Bar Plot from a dataframe that is reordered in descending order. I would like the bar name to be the index (0 to 5) When I plot my Plot, it does not appear in descending order. How can I change the order of the bars?

Operations I’ve already tried:

import pandas as pd
import matplotlib.pyplot as plt 
plt.style.use('seaborn')
des = ['0. item A', '1. item B', '2. item C', '3. item D','4. item E','5. item F']
total = [100,75,5,30,3,5]

df = pd.DataFrame({'des': des,
                   'total':total}) 
df = df.sort_values(by=['total'], ascending=False)

df

# Solução 1
xs = df.index
ys = df['total']

plt.bar(xs, ys, color='#DDA63A')

# Solução 2:
plt.bar('index', 'total', color='#DDA63A', data=df)

#Solução 3
plt.bar(df.index, df.total, color='#DDA63A', data=df)

Answer 1

One solution is not to use Matplotlib, but the Plot of the Pandas Dataframe itself. You can sort the Dataframe and plot as follows:

df.sort_values('total', ascending=False)[['des','total']].plot.bar();

For more similar answers, in English:
https://stackoverflow.com/questions/40902958/pandas-dataframe-bar-plot-with-sort-values-by-other-column

Answer 2

One of the ways you can get the desired result is:

import matplotlib.pyplot as plt
plt.style.use('seaborn')
des = ['0. item A', '1. item B', '2. item C', '3. item D','4. item E','5. item F']
total = [100,75,5,30,3,5]

df = pd.DataFrame({'des': des,
                   'total':total})
df = df.sort_values(by=['total'], ascending=False)

xs = df.index
ys = df['total']

plt.bar(sorted(xs), ys, color='#DDA63A')
plt.show()

Note that the only change made was in the penultimate line reordering the variable Xs with the function Sorted().

Answer 3

Convert the index to a string list, it is a way to circumvent the captive matplotlib ordering.

`xs = df.index.map(str)
 ys = df['total']
 plt.bar(xs, ys, color='#DDA63A')`