In [78]:

    

from __future__ import division
from __future__ import print_function
from math import log


    

In [1]:

    

def createDataset():
    dataSet = [[1,1,'yes'],[1,1,'yes'],[1,0,'no'],[0,1,'no'],[0,1,'no']]
    labels = ['no surfacing', 'flippers'] 
    return dataSet, labels


    

In [2]:

    

data, labels = createDataset()


    

In [3]:

    

data


    

    
Out[3]:





[[1, 1, 'yes'], [1, 1, 'yes'], [1, 0, 'no'], [0, 1, 'no'], [0, 1, 'no']]

In [18]:

    

x = {'a':1,'b':2}


    

In [25]:

    

x['c'] = 3


    

In [33]:

    

x['c'] += 1


    

In [34]:

    

x


    

    
Out[34]:





{'a': 1, 'b': 2, 'c': 5}

In [81]:

    

def calcEnt(data, pos):
    numrows = len(data)
    count = {}
    entropy = 0
    
    # freq distrbution 
    for i in data:
        key = i[pos]
        if key not in count.keys():
            count[key] = 1
        else:
            count[key] += 1
    # entropy
    for key in count.keys():
        c = count[key]
        p = c / numrows 
        ent = p * log(p, 2)
        entropy -= ent
            
    return entropy


    

In [82]:

    

calcEnt(data,2)


    

    
Out[82]:





0.9709505944546686

In [88]:

    

def infGain(data, target, var):
    numrows = len(data)
    count = {}
    entropy = 0
    
    # freq distrbution 
    for i in data:
        key = i[var]
        if key not in count.keys():
            count[key] = 1
        else:
            count[key] += 1
    
    for key in count.keys():
        subSet = []
        for i in data:
            if i[var] == key:
                subSet.append(i)
        print(subSet)
        ent = calcEnt(subSet, target)
        print (ent)


    

In [91]:

    

infGain(data, 2,1)


    

    




[[1, 0, 'no']]
0.0
[[1, 1, 'yes'], [1, 1, 'yes'], [0, 1, 'no'], [0, 1, 'no']]
1.0

In [ ]: