How can I add the counts to the histogram plot?

Question

I want to add the counts data of histogram to the plot in matplotlib. Here is my data;

import matplotlib.pyplot as plt
data = [['B', 1], ['C', 2], ['A', 3],['A', 4], ['B', 5], ['B', 6],['A', 7], ['B', 8], ['C', 9],['D',10]] 
df = pd.DataFrame(data, columns = ['Name', 'Count']) 
plt.hist(df['Name'])
plt.show()

The result is like this; result1

I tried to use plt.text and value_counts() but their sorting are different...

import matplotlib.pyplot as plt
data = [['B', 1], ['C', 2], ['A', 3],['A', 4], ['B', 5], ['B', 6],['A', 7], ['B', 8], ['C', 9],['D',10]] 
df = pd.DataFrame(data, columns = ['Name', 'Count']) 

xvals = np.arange(len(df['Name'].value_counts()))
yvals = list(df['Name'].value_counts())
for i in range(len(df['Name'].value_counts())):
    plt.text(x=xvals[i], y=yvals[i],s=df['Name'].value_counts(sort=False)[i])

plt.hist(df['Name'])
plt.show()

So, I get a result like this; result2

I think it mustn't be so difficult but I can't find any solution.

Check out this answer: https://stackoverflow.com/a/30229062/2943652 — Francisca Concha-Ramírez, Nov 22 '19 at 13:27
thanks but, it is bar chart, not histogram... So, can't solve the problem — msklc, Nov 22 '19 at 13:57
you can do the same thing for histogram. I will post an answer. — Francisca Concha-Ramírez, Nov 22 '19 at 14:06

score 4 · Accepted Answer · answered Nov 22 '19 at 13:40

You can try something like this:

hist returns counts, bins and patches. patches is a list of rectangles. Then you can annotate the axis using the count and coordinates of the patch rectangles.

import numpy as np
import matplotlib.pyplot as plt

# generate some random data
x = np.random.randn(10000)
x = x * 100
x = x.astype(np.int)

# plot the histogram of the data
bins = np.arange(-300,300,20)
fig = plt.figure(figsize=(15,4))
ax = plt.gca()
counts, _, patches = ax.hist(x, bins=bins,edgecolor='r')
for count, patch in zip(counts,patches):
    ax.annotate(str(int(count)), xy=(patch.get_x(), patch.get_height()))
plt.show()

score 2 · Answer 2 · answered Nov 22 '19 at 14:10

pyplot.hist returns the lengths and locations of the bins. You can make use of this by saving the returns from pyplot.hist:

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

data = [['B', 1], ['C', 2], ['A', 3],['A', 4], ['B', 5], ['B', 6],['A', 7], ['B', 8], ['C', 9],['D',10]] 
df = pd.DataFrame(data, columns = ['Name', 'Count']) 

xvals = np.arange(len(df['Name'].value_counts()))
yvals = list(df['Name'].value_counts())

counts, bins, _ = plt.hist(df['Name'])

for n, b in zip(counts, bins):
        plt.gca().text(b + 0.1, n, str(n))  # +0.1 to center text

plt.show()

This results in:

If you want to remove the 0.0s of the empty bars, change the for loop to this:

for n, b in zip(counts, bins):
    if n > 0:
        plt.gca().text(b + 0.1, n, str(n))  # +0.1 to center text

Result:

How can I add the counts to the histogram plot?

2 Answers2

Linked