Reputation: 724
I have a pandas
dataframe similar to the following
names
x 3.5
x 3.7
z 2.8
x 3.4
y 3.25
z 2.9
...
And I wish to make a comparative boxplot (three boxplots next to each other for each of x
, y
, and z
. I'm using the seaborn
package, and I can only get a boxplot for all of the values combined. What am I doing wrong?
b = sns.boxplot(data = dat);
Upvotes: 0
Views: 2570
Reputation: 7150
I think you can draw the side by side boxplot this way:
import pandas as pd
import seaborn as sns
from io import StringIO
data = """
names,num
x,3.5
x,3.7
z,2.8
x,3.4
y,3.25
z,2.9
"""
df = pd.read_csv(StringIO(data), header=0)
to_replace = {0:'x', 1:'y', 2:'z'}
df['names'] = df['names'].replace(to_replace=to_replace)
order = ["x", "y", "z"]
sns.boxplot(x="names", y="num", data=df, order=order)
Here is the boxplot:
Refs:
Upvotes: 1