How to get matplotlib bar chart to match numeric count in python terminal

Question

My main objective is to be consistent with both my numeric output and my visual output. However, I can't seem to get to them to match.

Here is my setup using python 3.x:

df = pd.DataFrame([ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1,
   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1,
   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2],columns=['Expo'])

Followed by my setup for the bar chart in matplotlib:

x = df['Expo']
N = len(x)
y = range(N)
width = 0.125
plt.bar(x, y, width, color="blue")
fig = plt.gcf();

A Nice pretty graph produced:

However, using this snippet code to check and see what the actual numeric counts of both classes are...

print("Class 1: "+str(df['Expo'].value_counts()[1]),"Class 2: "+str(df['Expo'].value_counts()[2]))

I get the below:

Class 1: 85 Class 2: 70

Since I have 155 records in the data frame, numerically this makes sense. Having a single bar in the bar chart be at 155 does not.

I appreciate any help in advance.

Explain yourself better, because I see that the results you get are correct. — eyllanesc
– eyllanesc, Commented Jul 4, 2017 at 19:46
The numeric output for class 1 shows 85. However, the graph shows approximately 115. The numeric output for class 2 shows 70. However, the graph shows approximately 155. I want the numeric values and the graphical values to be the same. Currently, they don't seem to match. — Student
– Student, Commented Jul 4, 2017 at 19:48

Gabriel · Accepted Answer · 2017-07-04 19:52:45Z

I guess something like this is what you're after:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt


df = pd.DataFrame([ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1,
   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1,
   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
   1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2],columns=['Expo'])

# Count number of '1' and '2' elements in df
N1, N2 = len(df[df['Expo'] == 1]), len(df[df['Expo'] == 2])
width = 0.125
# Plot the lengths in x positions [1, 2]
plt.bar([1, 2], [N1, N2], width, color="blue")
fig = plt.gcf()
plt.show()

Which produces

ImportanceOfBeingErnest · Accepted Answer · 2017-07-04 20:12:16Z

1

You may use a histogram,

plt.hist(df["Expo"])

or specifying the bins

plt.hist(df["Expo"], bins=[0.5,1.5,2.5], ec="k")
plt.xticks([1,2])

answered Jul 4, 2017 at 20:12

ImportanceOfBeingErnest

342k61 gold badges737 silver badges771 bronze badges

Collectives™ on Stack Overflow

How to get matplotlib bar chart to match numeric count in python terminal

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related