In [1]:

    

import pandas as pd
import numpy as np


    

In [ ]:

    

iris = pd.read_csv("http://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data",
                  header = False)
print iris.shape
print iris.columns
print iris.head()
print iris.describe


    

In [17]:

    

# renaming columns
iris.columns = ["Sepal.Length", "Sepal.Width", "Petal.Length", "Petal.Width", "Species"]


    

In [49]:

    

# normalization function
def normalize(x):
    num = x - np.min(x)
    denom = np.max(x) - np.min(x)
    return (num / denom)


    

In [52]:

    

# normalization
iris_subset = iris.iloc[:, 1:4]
iris_norm = (iris_subset - np.min(iris_subset)) / (np.max(iris_subset) - np.min(iris_subset))

print iris_subset.head()
print iris_norm.head()


    

    




   Sepal.Width  Petal.Length  Petal.Width
0          3.0           1.4          0.2
1          3.2           1.3          0.2
2          3.1           1.5          0.2
3          3.6           1.4          0.2
4          3.9           1.7          0.4
   Sepal.Width  Petal.Length  Petal.Width
0     0.416667      0.067797     0.041667
1     0.500000      0.050847     0.041667
2     0.458333      0.084746     0.041667
3     0.666667      0.067797     0.041667
4     0.791667      0.118644     0.125000

In [57]:

    

# Train test split
from sklearn.cross_validation import train_test_split

iris_features = iris_norm
iris_labels = iris['Species']

features_train, features_test, labels_train, labels_test = train_test_split(
    iris_features, iris_labels, test_size = 0.33, random_state = 77)


    

In [66]:

    

# Machine learning
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors = 3)
knn.fit(features_train, labels_train)
predictions = knn.predict(features_test)


    

    
Out[66]:





array(['Iris-virginica', 'Iris-virginica', 'Iris-virginica',
       'Iris-versicolor', 'Iris-setosa', 'Iris-virginica',
       'Iris-virginica', 'Iris-virginica', 'Iris-setosa', 'Iris-versicolor'], dtype=object)

In [69]:

    

# Model evaluation
from sklearn.metrics import accuracy_score

accuracy_score(labels_test, predictions)


    

    
Out[69]:





0.95999999999999996