Seaborn Scatterplot matrix / pairplot integration

[scls19fr]

import seaborn as sns
sns.set()

df = sns.load_dataset("iris")
sns.pairplot(df, hue="species")

displays

but pairplot doesn't work the same way with ModelFrame

import pandas as pd
pd.set_option('max_rows', 10)
import sklearn.datasets as datasets
import pandas_ml as pdml  # https://github.com/pandas-ml/pandas-ml
import seaborn as sns
import matplotlib.pyplot as plt
df = pdml.ModelFrame(datasets.load_iris())
sns.pairplot(df, hue=".target")

There is some useless subplots

[sinhrks]

Both behaviors are valid. The difference is caused by its dtype.

Seaborn

Species columns is object. If it is changed to numeric, the column will be drawn.

df = sns.load_dataset("iris")

species, _ = df['species'].factorize()
df['species'] = species
sns.pairplot(df, hue="species")
# species column is included

pdml (scikit-learn)

Species columns is int. If it is changed to object, the column will not be drawn.

df = pdml.ModelFrame(datasets.load_iris())
df['.target'] = df['.target'].apply(lambda x: 'Species{0}'.format(x))
sns.pairplot(df, hue=".target")
# target column is excluded

[scls19fr]

Ok thanks