In [1]:

    

# Maybe they found that very colorful shoes have done well the past.

data = []

for i in range(5000):
    
    if rd.random() <= .35:
        bought = 'N'
    else:
        bought = 'Y'
    
    if bought == 'Y':
        if rd.random() <= .85:
            shoes_has_multiple_colors = 1
        else:
            shoes_has_multiple_colors = 0
        if rd.random() <= .75:
            if rd.random() <= .25:
                r_times = rd.randint(4,5)
            else:
                r_times = rd.randint(1,5)
        else:
            r_times = 0
        if rd.random() <= .35:
            shoes_has_cool_features = 1
        else:
            shoes_has_cool_features = 0
    else:
        if rd.random() <= .35:
            shoes_has_multiple_colors = 1
        else:
            shoes_has_multiple_colors = 0
        if rd.random() <= .15:
            shoes_has_cool_features = 1
        else:
            shoes_has_cool_features = 0
        r_times = rd.randint(0,2)
    
    data.append([i, r_times, shoes_has_cool_features, shoes_has_multiple_colors, bought])

df = pd.DataFrame(data, columns=['id', 'run-times', 'shoe-has-cool-features', 'shoe-has-colors', 'bought'])
# df.to_csv('csv_output/ch17_fig1.csv', index=False)
df = pd.read_csv('csv_output/ch17_fig1.csv')
df.head()


    

    
Out[1]:






  

    

      

      
id
      
run-times
      
shoe-has-cool-features
      
shoe-has-colors
      
bought
    
  
  

    

      
0
      
0
      
2
      
0
      
0
      
N
    
    

      
1
      
1
      
1
      
0
      
1
      
N
    
    

      
2
      
2
      
0
      
0
      
0
      
N
    
    

      
3
      
3
      
0
      
0
      
1
      
N
    
    

      
4
      
4
      
5
      
0
      
1
      
Y
    
  

In [3]:

    

df = pd.read_csv('csv_output/ch17_fig1.csv')

%matplotlib inline
sns.set_style("white")
cm = sns.color_palette('Blues', 6)

# f, ax = plt.subplots(1,1, figsize=(8,8))

from sklearn.externals.six import StringIO
from sklearn import tree
import pydot
from IPython.display import Image 
clf = tree.DecisionTreeClassifier()
clf = clf.fit(df.iloc[:,1:-1], df.iloc[:,-1])


with open("svg_output/ch17_fig1.dot", 'w') as f:
    f = tree.export_graphviz(clf, out_file=f)

dot_data = StringIO()
tree.export_graphviz(clf, out_file=dot_data,  
                         feature_names=df.columns[1:-1],  
                         class_names=['Y', 'N'],  
                         filled=True, rounded=True,  
                         special_characters=True)  
graph = pydot.graph_from_dot_data(dot_data.getvalue())
# graph[0].write_svg('svg_output/ch17_fig1.svg') # write to svg
Image(graph[0].create_png())


    

    
Out[3]: