Boxplot with different y-axes and different y-scales in seaborn

Question

I am trying to create a boxplot with different y-axes and y-scales in seaborn but got stuck here.

In matplotlib I can use the following code to obtain my result:

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


# create random dataframe with different scales
df = pd.DataFrame(np.random.rand(30, 5), columns=['A', 'B', 'C', 'D', 'E'])
df['A'] *= 5
df['C'] *= 10
df['E'] *= 15


# create boxplot with a different y scale for different rows
selection = ['A', 'C', 'E']
fig, ax = plt.subplots(1, len(selection), figsize=(10, len(selection)))
i = 0
for col in selection:
    axo = df[col].plot(kind='box', ax=ax[i], showfliers=False, grid=True)
    axo.set_ylim(df[col].min(), df[col].max())
    axo.set_ylabel(col + ' / Unit')
    i += 1

plt.tight_layout()
plt.show()

which yields:

I have tried to replace the actual boxplot using a seaborn boxplot but it did not work.

Now my question is: How can I obtain the same plotting result using seaborn?

Thanks in advance for your help!

Cheers Cord

Answer 1

This is a code snippet I am using in my notebook, Can try using it this way.

columns = df_plt.columns # my data has 8 columns.

fig, ax = plt.subplots(ncols = 7, figsize=(19,10))
plt.subplots_adjust(wspace = 1)
sns.set_palette("PRGn")
for i in range(0,7):
    s = sns.boxplot(ax = ax[i], data = df_plt[columns[i]], showfliers = False)

    ax[i].set_xlabel(columns[i])

plt.show()

Answer 2

I found a solution on my own:

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


# create random dataframe with different scales
df = pd.DataFrame(np.random.rand(30, 5), columns=['A', 'B', 'C', 'D', 'E'])
df['A'] *= 5
df['C'] *= 10
df['E'] *= 15

# create boxplot with a different y scale for different rows
selection = ['A', 'C', 'E']
fig, axes = plt.subplots(1, len(selection))
for i, col in enumerate(selection):
    ax = sns.boxplot(y=df[col], ax=axes.flatten()[i])
    ax.set_ylim(df[col].min(), df[col].max())
    ax.set_ylabel(col + ' / Unit')
plt.show()

yields:

Thanks anyway!