In [1]:

    

# If you see a numpy error because of numpy.dtype changed, fix it by installing pandas by hand with the following command
# pip install --no-binary pandas -I pandas
import seaborn as sns
import pandas as pd


    

In [2]:

    

# Load dataset
# This doesn't work, SSL error
# iris = sns.load_dataset('iris')
# I'm using a submodule for the seaborn data
iris = pd.read_csv('seaborn-data/iris.csv')


    

In [3]:

    

iris.head()


    

    
Out[3]:







  

    

      

      
sepal_length
      
sepal_width
      
petal_length
      
petal_width
      
species
    
  
  

    

      
0
      
5.1
      
3.5
      
1.4
      
0.2
      
setosa
    
    

      
1
      
4.9
      
3.0
      
1.4
      
0.2
      
setosa
    
    

      
2
      
4.7
      
3.2
      
1.3
      
0.2
      
setosa
    
    

      
3
      
4.6
      
3.1
      
1.5
      
0.2
      
setosa
    
    

      
4
      
5.0
      
3.6
      
1.4
      
0.2
      
setosa
    
  

In [4]:

    

# Create the inputs with only the features
X_iris = iris.drop('species', axis=1)


    

In [6]:

    

# Show the shape of the input vectors
X_iris.shape


    

    
Out[6]:





(150, 4)

In [7]:

    

# Create the labels that we want to predict
Y_iris = iris['species']


    

In [8]:

    

# Show the shape of the label vector
Y_iris.shape


    

    
Out[8]:





(150,)

In [10]:

    

# Scikit-Learn Estimator API
from sklearn.naive_bayes import GaussianNB # 1. Choose model class
# Instantiate the model
model = GaussianNB()
# Fit the model
model.fit(X_iris, Y_iris)


    

    
Out[10]:





GaussianNB(priors=None)

In [ ]: