20BECE30058

from google.colab import drive

drive.mount('/content/drive')

Mounted at /content/drive

import pandas as pd
import numpy as np

df=pd.read_csv('/content/drive/MyDrive/ML Lab/HCP/Classified Data',index_col=0)

print(df.head()) from sklearn.model_selection import train_test_split

X=df.drop('TARGET
CLASS',axis=1) 2

0 0.643798 0.879422 1.231409 2

2 1.154483 0.957877 1.285597 0 3 1.380003 1.522692 1.153093 2
4 0.646691 1.463812 1.419167 2

y=df['TARGET CLASS']

X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.2,random_state=100)

WTT PTI EQW SBI LQE QWG FDJ

\ 0 0.913917 1.162073 0.567946 0.755464 0.780862 0.352608 0.759697
1 0.635632 1.003722 0.535342 0.825645 0.924109 0.648450 0.675334
2 0.721360 1.201493 0.921990 0.855595 1.526629 0.720781 1.626351
3 1.234204 1.386726 0.653046 0.825624 1.142504 0.875128 1.409708
4 1.279491 0.949750 0.627280 0.668976 1.232537 0.703727 1.115596

PJF HQE NXJ TARGET

CLASS 1 1.013546 0.621552 1.492702
from sklearn.neighbors 0
import KNeighborsClassifier
knn=KNeighborsClassifier(n_neighbors=1)
knn.fit(X_train,y_train) pred=knn.predict(X_test)

----Arguments----

KNeighborsClassi er(

n_neighbors=5,

weights='uniform'(---'uniform' or 'callable'),

algorithm='auto'({'auto', 'ball_tree', 'kd_tree', 'brute'},Algorithm used to compute the nearest neighbors),

leaf_size=30,

p=2(---Power parameter. When p = 1, this is

equivalent to using manhattan_distance (l1), and euclidean_distance (l2) for p = 2),

metric='minkowski',

metric_params=None(---the distance metric to use for the tree),

n_jobs=None,

from sklearn.metrics import classification_report,confusion_matrix

print(confusion_matrix(y_test,pred))
print(classification_report(y_test,pred))

[[98 12]
[ 8 82]]
precision recall f1-score support

0 0.92 0.89 0.91 110

1 0.87 0.91 0.89 90
accuracy 0.90 200
macro avg 0.90 0.90 0.90 200
weighted avg 0.90 0.90 0.90 200

1
 20BECE30058
KNN using Standard Scaler

1) Split the Dataset

#------Here we are not knowing that what are the features so how to group the data points?
#-----If the values of some features are higher than it is required to do the feature scaling otherwise such features will show a very majo
#-----it will have much effect on the distance between the features

import pandas as pd
import numpy as np

df=pd.read_csv('/content/drive/MyDrive/ML Lab/HCP/Classified Data',index_col=0)

print(df.head()) from sklearn.model_selection import train_test_split

X=df.drop('TARGET CLASS',axis=1)

y=df['TARGET CLASS']

X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.2,random_state=100)

WTT PTI EQW SBI LQE QWG FDJ

\ 0 0.913917 1.162073 0.567946 0.755464 0.780862 0.352608 0.759697
1 0.635632 1.003722 0.535342 0.825645 0.924109 0.648450 0.675334
2 0.721360 1.201493 0.921990 0.855595 1.526629 0.720781 1.626351
3 1.234204 1.386726 0.653046 0.825624 1.142504 0.875128 1.409708
4 1.279491 0.949750 0.627280 0.668976 1.232537 0.703727 1.115596

PJF HQE NXJ TARGET

CLASS 0 0.643798 0.879422 1.231409 1
1 1.013546 0.621552 1.492702 0
2 1.154483 0.957877 1.285597 0
3 1.380003 1.522692 1.153093 1
4 0.646691 1.463812 1.419167 1

2) Scale the Splitted dataset

1st Fit the data

2nd Transform the data

from sklearn.preprocessing import StandardScaler

scaler=StandardScaler()
scaler.fit(X_train) #--- It will drop the target class as we dont want to scale the labels

StandardScaler()

scaled_features_X_train=scaler.transform(X_train)
scaled_features_X_test=scaler.transform(X_test)

3) Apply KNN Model on the scaled dataset

from sklearn.neighbors import KNeighborsClassifier

knn=KNeighborsClassifier(n_neighbors=1) #---means k=1

knn.fit(scaled_features_X_train,y_train)
pred_1=knn.predict(scaled_features_X_test) pred

array([0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 0,
1, 0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1,
1, 0, 0, 1, 1, 0, 0, 0, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 0,
0, 0, 1, 0, 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0,
0, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1,
1, 1, 0, 0, 1, 0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 1, 1, 1, 1, 0,
0, 0, 1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 0, 0,
1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 0, 1,
0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 0, 0, 0,
0, 1])

2
 20BECE30058
4) Find the Classi cation Report for KNN =1 using scaled Data

from sklearn.metrics import classification_report,confusion_matrix

print(confusion_matrix(y_test,pred_1))
print(classification_report(y_test,pred_1))

#---Here you can see that the number of Misclassifications(17) in scaled dataset is more as compared to unscaled dataset(20).

[[98 12] [ 5 85]]

precisionrecall f1-scoresupport

0 0.95 0.89 0.92 110

1 0.88 0.94 0.91 90

accuracy 0.92 200

macro avg 0.91 0.92 0.91 200
weighted avg 0.92 0.92 0.92 200

'Elbow' method to nd correct value of 'k'

#------Use elbow method to choose correct value of k

#------Use the model with different values of 'k' and plot the error rate
#-----and observe which one has minimum error rate

error_rate=[] #-----empty list

for i in range(1,40):
knn=KNeighborsClassifier(n_neighbors=i)
knn.fit(scaled_features_X_train,y_train)
pred_i=knn.predict(scaled_features_X_test)
error_rate.append(np.mean(pred_i != y_test))
#-- - -taking the mean of all prediction and actual labels which are not equal

print(error_rat)

[0.085, 0.09, 0.09, 0.08, 0.09, 0.075, 0.09, 0.075, 0.095, 0.075, 0.095, 0.075, 0.08, 0.08, 0.075, 0.085, 0.085, 0.085, 0.08, 0.085

import matplotlib.pyplot as plt

plt.figure(figsize=(10,6))
plt.plot(range(1,40),error_rate,color='blue',linestyle='--',marker='o')
plt.title('Error Rate vs K value (1 to 40)') plt.xlabel('K value')
plt.ylabel('Error rate') plt.grid()

knn=KNeighborsClassifier(n_neighbors=32)

3
 20BECE30058
knn.fit(scaled_features_X_train,y_train) pred_28=knn.predict(scaled_features_X_test)

print(confusion_matrix(y_test,pred_28)) print('\n') print(classification_report(y_test,pred_28))

#---Compare the confusion matrix for k=1 and for k=28 , it has better classsification

#---Misclassifications without Scaled dataset : 20 #---Misclassifications with Scaled dataset :17

#---Misclassification with better 'k' value and scaled dataset :16

[[99 11]
[ 3 87]]

precision recall f1-score support

0 0.97 0.90 0.93 110

1 0.89 0.97 0.93 90

accuracy 0.93 200

macro avg 0.93 0.93 0.93 200
weighted avg 0.93 0.93 0.93 200