glass_knn.py

# -*- coding: utf-8 -*-
"""
Created on Fri Nov  6 15:45:17 2020

@author: HP
"""
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

glass=pd.read_csv('C:/Users/HP/Desktop/assignments submission/KNN/glass.csv')
 
#to split train and test data
from sklearn.model_selection import train_test_split
train,test=train_test_split(glass,test_size=0.3,random_state=0)

#KNN
from sklearn.neighbors import KNeighborsClassifier as KNC
#to find best k value
acc=[]
for i in range(3,50,2):
    neigh=KNC(n_neighbors=i)
    neigh.fit(train.iloc[:,0:9],train.iloc[:,9])
    train_acc=np.mean(neigh.predict(train.iloc[:,0:9])==train.iloc[:,9])
    test_acc=np.mean(neigh.predict(test.iloc[:,0:9])==test.iloc[:,9])
    acc.append([train_acc,test_acc])
    
plt.plot(np.arange(3,50,2),[i[0] for i in acc],'bo-')
plt.plot(np.arange(3,50,2),[i[1] for i in acc],'ro-')
plt.legend(['train','test'])


#from plots at k=5 we get best model
#model building at k=5 
neigh=KNC(n_neighbors=5)
neigh.fit(train.iloc[:,0:9],train.iloc[:,9])
pred_train=neigh.predict(train.iloc[:,0:9])
train_acc=np.mean(pred_train==train.iloc[:,9])
train_acc#0.76
pred_test=neigh.predict(test.iloc[:,0:9])
test_acc=np.mean(pred_test==test.iloc[:,9])
test_acc#0.661