Missing Values Visualizer Examples¶

In [1]:

%load_ext autoreload

%autoreload 2

import sys
sys.path.append("./../..")

In [2]:

%reload_ext yellowbrick
%matplotlib inline
# Imports
import pandas as pd  
import numpy as np
import matplotlib.pyplot as plt
import matplotlib
from sklearn.preprocessing import OneHotEncoder
from yellowbrick.contrib.missing import MissingValuesDispersion, MissingValuesBar
from sklearn.datasets import make_classification

Use the Horse Colic Data Set¶

Contains natural missing values in data

In [3]:

headers = pd.read_csv("./horse-colic.attrs")

In [4]:

url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/horse-colic/horse-colic.data'

# Retrieve Data Set
df = pd.read_csv(url, delim_whitespace=True)
df.columns = headers.Attribute.values
df.replace(to_replace="?", value=np.nan, inplace=True,)

In [5]:

X = df.drop(['cp_data'], axis=1)
y = df['cp_data']

Using Missing Values Dispersion Chart¶

In [6]:

classes=['sick', 'healthy']

No target y passed in, produces mono-color chat¶

In [7]:

viz = MissingValuesDispersion(classes=classes)
viz.fit(X)
viz.show()

No handles with labels found to put in legend.

Target y passed in, produces Dispersion chart with elements colored by target variable¶

In [8]:

viz = MissingValuesDispersion(classes=classes)
viz.fit(X, y=y)
viz.show()

Using Missing Values Bar Chart¶

No target y passed in, produces mono-color bar chart¶

In [9]:

oz = MissingValuesBar(classes=classes)
oz.fit(X)
oz.show()

No handles with labels found to put in legend.

Target y passed in, produces stacked bar chart¶

In [10]:

oz = MissingValuesBar(classes=classes)
oz.fit(X, y=y)
oz.show()

In [ ]: