How to create a grouped bar plot from lists

user_10755782 0
Viewed 443
  • I'm attempting to plot a bar plot, which compares the number of several items in 2 different cases.
  • The desired output would be a bar plot with 4+4 = 8 bars, next to to each other which indicates the number of each type for each case.
  • This is the initial code which I wrote, it is not giving what I expect. How can I modify this?
import numpy
import matplotlib.pyplot as plt

names = ["a","b","c","d"]
case1 = [5,7,5,6]
case2 = [7,4,8,5]

plt.hist(case1)
plt.show()
pandas bar-chart matplotlib plot python
3 Answers
  • The simplest way is to create a dataframe with pandas, and then plot with pandas.DataFrame.plot
    • The dataframe index, 'names' in this case, is automatically used for the x-axis and the columns are plotted as bars.
    • matplotlib is used as the plotting backend
  • Tested in python 3.8, pandas 1.3.1 and matplotlib 3.4.2
  • For lists of uneven length, see How to create a grouped bar plot from lists of uneven length
import pandas as pd
import matplotlib.pyplot as plt

names = ["a","b","c","d"]
case1 = [5,7,5,6]
case2 = [7,4,8,5]

# create the dataframe
df = pd.DataFrame({'c1': case1, 'c2': case2}, index=names)

# display(df)
   c1  c2
a   5   7
b   7   4
c   5   8
d   6   5

# plot
ax = df.plot(kind='bar', figsize=(6, 4), rot=0, title='Case Comparison', ylabel='Values')
plt.show()

enter image description here

  • Try the following for python 2.7
fig, ax = plt.subplots(figsize=(6, 4))
df.plot.bar(ax=ax, rot=0)
ax.set(ylabel='Values')
plt.show()
user_77588040

It can be achieved by adapting this code to your problem.

# importing pandas library
import pandas as pd
# import matplotlib library
import matplotlib.pyplot as plt
  
# creating dataframe
df = pd.DataFrame({
    'Names': ["a","b","c","d"],
    'Case1': [5,7,5,6],
    'Case2': [7,4,8,5]
})
  
# plotting graph
df.plot(x="Names", y=["Case1", "Case2"], kind="bar")
user_37968640

Matplotlib only (plus numpy.arange).

It's easy to place correctly the bar groups if you think about it.

enter image description here

import matplotlib.pyplot as plt
from numpy import arange

places = ["Nujiang Lisu","Chuxiong Yi","Liangshan Yi","Dehong Dai & Jingpo"]
animals = ['Pandas', 'Snow Leopards']

n_places = len(places)
n_animals = len(animals)

animals_in_place = [[5,7,5,6],[7,4,8,5]]

### prepare for grouping the bars    
total_width = 0.5 # 0 ≤ total_width ≤ 1
d = 0.1 # gap between bars, as a fraction of the bar width, 0 ≤ d ≤ ∞
width = total_width/(n_animals+(n_animals-1)*d)
offset = -total_width/2

### plot    
x = arange(n_places)
fig, ax = plt.subplots()
for animal, data in zip(animals, animals_in_place):
    ax.bar(x+offset, data, width, align='edge', label=animal)
    offset += (1+d)*width
ax.set_xticks(x) ; ax.set_xticklabels(places)
fig.legend()
user_27493970
Related