Here I will show how to add frequency count to the each category in the box plot. For this we will use the searborn provided example dataset.
First, lets load the dataset:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
import searborn as sns
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
matplotlib.rcParams['pdf.fonttype'] = 42
matplotlib.rcParams['ps.fonttype'] = 42
data = sns.load_dataset('tips')
tips.head()
total_bill tip sex smoker day time size
0 16.99 1.01 Female No Sun Dinner 2
1 10.34 1.66 Male No Sun Dinner 3
2 21.01 3.50 Male No Sun Dinner 3
3 23.68 3.31 Male No Sun Dinner 2
4 24.59 3.61 Female No Sun Dinner 4
After that, lets create two separate functions to get the all the annotation informations.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
def hue_annotation_counts(df, label, yaxis_val, hue_label, order=None, hue_order=None):
if order==None:
order = list(df[label].unique())
if hue_order == None:
hue_order = list(df[hue_label].unique())
width=0.8
n_levels = len(df[hue_label].unique())
each_width = width / n_levels
offsets = np.linspace(0, width - each_width, n_levels)
offsets -= offsets.mean()
pos = [x+o for x in np.arange(len(order)) for o in offsets]
counts = df.groupby([label,hue_label])[yaxis_val].size()
counts = counts.reindex(pd.MultiIndex.from_product([order, hue_order ]))
medians = df.groupby([label,hue_label])[yaxis_val].median()
medians = medians.reindex(pd.MultiIndex.from_product([order, hue_order]))
return pos, counts, medians, order, hue_order
def label_annotation_counts(df, label, yaxis_val, order=None):
if order==None:
order = list(df[label].unique())
counts = df.groupby(label)[yaxis_val].size()
counts = counts.reindex(index=order)
medians = df.groupby(label)[yaxis_val].median()
medians = medians.reindex(index=order)
return range(len(counts)), counts, medians, order
Now, plotting boxplots without hue information:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
plt.figure(figsize=(11,7))
pos, counts, medians, order = label_annotation_counts(temp, 'day','total_bill', order)
sns.boxplot(x = 'day', y='total_bill',data= temp, order=order)
for p,n,m in zip(pos,counts,medians):
if not np.isnan(m):
plt.annotate('N={:.0f}'.format(n), xy=(p, m), xycoords='data', ha='center', va='bottom', size=15)
plt.tight_layout()
plt.ylabel('Total Bill')
plt.xlabel('Day')
plt.xticks(rotation=90)
Now, plotting boxplots with extra information (‘sex’), which sub-divides x-label into further two categories. If you have your own order and hue order, you can pass that while calling the function.
1
2
3
4
5
6
7
8
9
10
11
12
plt.figure(figsize=(11,8))
pos, counts, medians, order, hue_order = hue_annotation_counts(temp, 'day','total_bill','sex')
sns.boxplot(x = 'day', y='total_bill',hue = 'sex',data= temp, order=order, hue_order=hue_order)
for p,n,m in zip(pos,counts,medians):
if not np.isnan(m):
plt.annotate('N={:.0f}'.format(n), xy=(p, m), xycoords='data', ha='center', va='bottom', size=10)
plt.tight_layout()
plt.xticks(rotation=90)
