Heart Disease Data¶

This is a multivariate type of dataset which means providing or involving a variety of separate mathematical or statistical variables, multivariate numerical data analysis.

It is composed of 14 attributes which are age, sex, chest pain type, resting blood pressure, serum cholesterol, fasting blood sugar, resting electrocardiographic results, maximum heart rate achieved,exercise-induced angina, oldpeak — ST depression induced by exercise relative to rest, the slope of the peak exercise ST segment, number of major vessels and Thalassemia.

This database includes 76 attributes, but all published studies relate to the use of a subset of 14 of them.

The Cleveland database is the only one used by ML researchers to date. One of the major tasks on this dataset is to predict based on the given attributes of a patient that whether that particular person has heart disease or not and other is the experimental task to diagnose and find out various insights from this dataset which could help in understanding the problem more.

In [399]:
import numpy as np
import pandas as pd
import os
import warnings
warnings.filterwarnings('ignore')
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn as sns
from sklearn import preprocessing
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import MinMaxScaler
from sklearn.decomposition import PCA
from sklearn.pipeline import Pipeline
In [400]:
path = 'C:\\Users\\mahit\\OneDrive\\Desktop\\DSPP\\ML & DL\\Assignments\\JNTUH ML DL assignment 3\\'
In [401]:
df = pd.read_csv(path+'heart_disease_uci.csv')
df
Out[401]:
id age sex dataset cp trestbps chol fbs restecg thalch exang oldpeak slope ca thal num
0 1 63 Male Cleveland typical angina 145.0 233.0 True lv hypertrophy 150.0 False 2.3 downsloping 0.0 fixed defect 0
1 2 67 Male Cleveland asymptomatic 160.0 286.0 False lv hypertrophy 108.0 True 1.5 flat 3.0 normal 2
2 3 67 Male Cleveland asymptomatic 120.0 229.0 False lv hypertrophy 129.0 True 2.6 flat 2.0 reversable defect 1
3 4 37 Male Cleveland non-anginal 130.0 250.0 False normal 187.0 False 3.5 downsloping 0.0 normal 0
4 5 41 Female Cleveland atypical angina 130.0 204.0 False lv hypertrophy 172.0 False 1.4 upsloping 0.0 normal 0
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
915 916 54 Female VA Long Beach asymptomatic 127.0 333.0 True st-t abnormality 154.0 False 0.0 NaN NaN NaN 1
916 917 62 Male VA Long Beach typical angina NaN 139.0 False st-t abnormality NaN NaN NaN NaN NaN NaN 0
917 918 55 Male VA Long Beach asymptomatic 122.0 223.0 True st-t abnormality 100.0 False 0.0 NaN NaN fixed defect 2
918 919 58 Male VA Long Beach asymptomatic NaN 385.0 True lv hypertrophy NaN NaN NaN NaN NaN NaN 0
919 920 62 Male VA Long Beach atypical angina 120.0 254.0 False lv hypertrophy 93.0 True 0.0 NaN NaN NaN 1

920 rows × 16 columns

In [402]:
df.head()
Out[402]:
id age sex dataset cp trestbps chol fbs restecg thalch exang oldpeak slope ca thal num
0 1 63 Male Cleveland typical angina 145.0 233.0 True lv hypertrophy 150.0 False 2.3 downsloping 0.0 fixed defect 0
1 2 67 Male Cleveland asymptomatic 160.0 286.0 False lv hypertrophy 108.0 True 1.5 flat 3.0 normal 2
2 3 67 Male Cleveland asymptomatic 120.0 229.0 False lv hypertrophy 129.0 True 2.6 flat 2.0 reversable defect 1
3 4 37 Male Cleveland non-anginal 130.0 250.0 False normal 187.0 False 3.5 downsloping 0.0 normal 0
4 5 41 Female Cleveland atypical angina 130.0 204.0 False lv hypertrophy 172.0 False 1.4 upsloping 0.0 normal 0
In [403]:
df.tail()
Out[403]:
id age sex dataset cp trestbps chol fbs restecg thalch exang oldpeak slope ca thal num
915 916 54 Female VA Long Beach asymptomatic 127.0 333.0 True st-t abnormality 154.0 False 0.0 NaN NaN NaN 1
916 917 62 Male VA Long Beach typical angina NaN 139.0 False st-t abnormality NaN NaN NaN NaN NaN NaN 0
917 918 55 Male VA Long Beach asymptomatic 122.0 223.0 True st-t abnormality 100.0 False 0.0 NaN NaN fixed defect 2
918 919 58 Male VA Long Beach asymptomatic NaN 385.0 True lv hypertrophy NaN NaN NaN NaN NaN NaN 0
919 920 62 Male VA Long Beach atypical angina 120.0 254.0 False lv hypertrophy 93.0 True 0.0 NaN NaN NaN 1
In [404]:
df = df.drop(['dataset'],axis = 1)
In [405]:
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 920 entries, 0 to 919
Data columns (total 15 columns):
 #   Column    Non-Null Count  Dtype  
---  ------    --------------  -----  
 0   id        920 non-null    int64  
 1   age       920 non-null    int64  
 2   sex       920 non-null    object 
 3   cp        920 non-null    object 
 4   trestbps  861 non-null    float64
 5   chol      890 non-null    float64
 6   fbs       830 non-null    object 
 7   restecg   918 non-null    object 
 8   thalch    865 non-null    float64
 9   exang     865 non-null    object 
 10  oldpeak   858 non-null    float64
 11  slope     611 non-null    object 
 12  ca        309 non-null    float64
 13  thal      434 non-null    object 
 14  num       920 non-null    int64  
dtypes: float64(5), int64(3), object(7)
memory usage: 107.9+ KB
In [406]:
df.isnull().sum()
Out[406]:
id            0
age           0
sex           0
cp            0
trestbps     59
chol         30
fbs          90
restecg       2
thalch       55
exang        55
oldpeak      62
slope       309
ca          611
thal        486
num           0
dtype: int64
In [407]:
df.describe()
Out[407]:
id age trestbps chol thalch oldpeak ca num
count 920.000000 920.000000 861.000000 890.000000 865.000000 858.000000 309.000000 920.000000
mean 460.500000 53.510870 132.132404 199.130337 137.545665 0.878788 0.676375 0.995652
std 265.725422 9.424685 19.066070 110.780810 25.926276 1.091226 0.935653 1.142693
min 1.000000 28.000000 0.000000 0.000000 60.000000 -2.600000 0.000000 0.000000
25% 230.750000 47.000000 120.000000 175.000000 120.000000 0.000000 0.000000 0.000000
50% 460.500000 54.000000 130.000000 223.000000 140.000000 0.500000 0.000000 1.000000
75% 690.250000 60.000000 140.000000 268.000000 157.000000 1.500000 1.000000 2.000000
max 920.000000 77.000000 200.000000 603.000000 202.000000 6.200000 3.000000 4.000000
In [408]:
df['cp'].value_counts()
Out[408]:
asymptomatic       496
non-anginal        204
atypical angina    174
typical angina      46
Name: cp, dtype: int64

Data Cleaning¶

In [409]:
fbs_1 = {'True':1, 'False':0}
df['fbs'] = df['fbs'].replace(fbs_1)

cp_1 = {'asymptomatic':0, 'non-anginal':1, 'atypical angina': 2, 'typical angina': 3}
df['cp'] = df['cp'].replace(cp_1)

exang_1 = {'True':1, 'False':0}
df['exang'] = df['exang'].replace(exang_1)

slope_1 = {'flat':0, 'upsloping':1, 'downsloping':2}
df['slope'] = df['slope'].replace(slope_1)

ca_1 = {'0.935653':4, '0.000000':0, '1.000000':1, '2.000000':2, '3.000000':3}
df['ca'] = df['ca'].replace(ca_1)

restecg_1 = {'normal':0, 'lv hypertrophy':1, 'st-t abnormality': 2}
df['restecg'] = df['restecg'].replace(restecg_1)

thal_1 = {'normal':0, 'reversable defect':1, 'fixed defect': 2}
df['thal'] = df['thal'].replace(thal_1)

sex_1 = {'Male':1, 'Female':0}
df['sex'] = df['sex'].replace(sex_1)
In [410]:
df['num'].replace([0,1,2,3,4],['absent','present','present','present','present'], inplace = True)
In [411]:
df['num'].value_counts(normalize = True)
Out[411]:
present    0.553261
absent     0.446739
Name: num, dtype: float64
In [412]:
df['num'].replace(['present','absent'],[1,0], inplace = True)
In [413]:
df['num'].value_counts(normalize = True)
Out[413]:
1    0.553261
0    0.446739
Name: num, dtype: float64

Data Imputation¶

In [414]:
# Numerical
df['trestbps'].fillna(df['trestbps'].mean(),inplace = True)
df['chol'].fillna(df['chol'].mean(),inplace = True)
df['thalch'].fillna(df['thalch'].mean(),inplace = True)
df['oldpeak'].fillna(df['oldpeak'].mean(),inplace = True)
df['ca'].fillna(df['ca'].std(),inplace = True)

# Categorical
df['exang'].fillna(df['exang'].mean(),inplace = True)
df['fbs'].fillna(df['fbs'].mean(),inplace = True)
df['slope'].fillna(df['slope'].mean(),inplace = True)
df['restecg'].fillna(df['restecg'].mean(),inplace = True)
df['thal'].fillna(df['thal'].mean(),inplace = True)
In [415]:
df.isnull().sum()
Out[415]:
id          0
age         0
sex         0
cp          0
trestbps    0
chol        0
fbs         0
restecg     0
thalch      0
exang       0
oldpeak     0
slope       0
ca          0
thal        0
num         0
dtype: int64
In [416]:
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 920 entries, 0 to 919
Data columns (total 15 columns):
 #   Column    Non-Null Count  Dtype  
---  ------    --------------  -----  
 0   id        920 non-null    int64  
 1   age       920 non-null    int64  
 2   sex       920 non-null    int64  
 3   cp        920 non-null    int64  
 4   trestbps  920 non-null    float64
 5   chol      920 non-null    float64
 6   fbs       920 non-null    object 
 7   restecg   920 non-null    float64
 8   thalch    920 non-null    float64
 9   exang     920 non-null    object 
 10  oldpeak   920 non-null    float64
 11  slope     920 non-null    float64
 12  ca        920 non-null    float64
 13  thal      920 non-null    float64
 14  num       920 non-null    int64  
dtypes: float64(8), int64(5), object(2)
memory usage: 107.9+ KB
In [417]:
df.describe()
Out[417]:
id age sex cp trestbps chol restecg thalch oldpeak slope ca thal num
count 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000 920.000000
mean 460.500000 53.510870 0.789130 0.750000 132.132404 199.130337 0.594771 137.545665 0.878788 0.538462 0.848570 0.654378 0.553261
std 265.725422 9.424685 0.408148 0.930969 18.443895 108.957634 0.793921 25.138494 1.053774 0.549850 0.555351 0.454881 0.497426
min 1.000000 28.000000 0.000000 0.000000 0.000000 0.000000 0.000000 60.000000 -2.600000 0.000000 0.000000 0.000000 0.000000
25% 230.750000 47.000000 1.000000 0.000000 120.000000 177.750000 0.000000 120.000000 0.000000 0.000000 0.935653 0.654378 0.000000
50% 460.500000 54.000000 1.000000 0.000000 130.000000 221.000000 0.000000 138.000000 0.800000 0.538462 0.935653 0.654378 1.000000
75% 690.250000 60.000000 1.000000 1.000000 140.000000 267.000000 1.000000 156.000000 1.500000 1.000000 0.935653 1.000000 1.000000
max 920.000000 77.000000 1.000000 3.000000 200.000000 603.000000 2.000000 202.000000 6.200000 2.000000 3.000000 2.000000 1.000000

Data Analysis¶

In [418]:
sns.displot(df['sex'])
Out[418]:
<seaborn.axisgrid.FacetGrid at 0x2062e4e3a60>
In [419]:
sns.displot(df['cp'])
Out[419]:
<seaborn.axisgrid.FacetGrid at 0x20639f05580>
In [420]:
sns.displot(df['age'])
Out[420]:
<seaborn.axisgrid.FacetGrid at 0x206354b7220>
In [421]:
sns.displot(df['thal'])
Out[421]:
<seaborn.axisgrid.FacetGrid at 0x20639f2f850>
In [422]:
sns.displot(df['restecg'])
Out[422]:
<seaborn.axisgrid.FacetGrid at 0x2063af85e20>
In [423]:
sns.displot(df['slope'])
Out[423]:
<seaborn.axisgrid.FacetGrid at 0x2063b0b6d90>
In [424]:
df.hist(column = "num", by = "sex", bins = 10)
Out[424]:
array([<AxesSubplot:title={'center':'0'}>,
       <AxesSubplot:title={'center':'1'}>], dtype=object)
In [425]:
plt.figure(figsize=(15,15))
cor = df.corr()
sns.heatmap(cor, annot=True, cmap=plt.cm.Reds)
Out[425]:
<AxesSubplot:>

Model Building¶

In [426]:
x = df.iloc[:,0:14]
y = df.iloc[:,-1]
In [427]:
x.head()
Out[427]:
id age sex cp trestbps chol fbs restecg thalch exang oldpeak slope ca thal
0 1 63 1 3 145.0 233.0 True 1.0 150.0 False 2.3 2.0 0.0 2.0
1 2 67 1 0 160.0 286.0 False 1.0 108.0 True 1.5 0.0 3.0 0.0
2 3 67 1 0 120.0 229.0 False 1.0 129.0 True 2.6 0.0 2.0 1.0
3 4 37 1 1 130.0 250.0 False 0.0 187.0 False 3.5 2.0 0.0 0.0
4 5 41 0 2 130.0 204.0 False 1.0 172.0 False 1.4 1.0 0.0 0.0
In [428]:
y.head()
Out[428]:
0    0
1    1
2    1
3    0
4    0
Name: num, dtype: int64
In [429]:
minmax=preprocessing.MinMaxScaler(feature_range=(0,1))
minmax.fit(x).transform(x)
Out[429]:
array([[0.        , 0.71428571, 1.        , ..., 1.        , 0.        ,
        1.        ],
       [0.00108814, 0.79591837, 1.        , ..., 0.        , 1.        ,
        0.        ],
       [0.00217628, 0.79591837, 1.        , ..., 0.        , 0.66666667,
        0.5       ],
       ...,
       [0.99782372, 0.55102041, 1.        , ..., 0.26923077, 0.31188434,
        1.        ],
       [0.99891186, 0.6122449 , 1.        , ..., 0.26923077, 0.31188434,
        0.32718894],
       [1.        , 0.69387755, 1.        , ..., 0.26923077, 0.31188434,
        0.32718894]])
In [430]:
x_train, x_test, y_train, y_test = train_test_split(x,y,train_size=0.8, random_state = 42, shuffle = True)
In [431]:
x_train.head()
Out[431]:
id age sex cp trestbps chol fbs restecg thalch exang oldpeak slope ca thal
880 881 62 1 0 132.132404 170.000000 False 2.0 120.000000 True 3.000000 0.538462 0.935653 0.654378
457 458 54 1 1 150.000000 199.130337 False 0.0 122.000000 False 0.000000 0.538462 0.935653 0.654378
797 798 51 1 1 132.132404 339.000000 False 0.0 137.545665 0.389595 0.878788 0.538462 0.935653 0.654378
25 26 50 0 1 120.000000 219.000000 False 0.0 158.000000 False 1.600000 0.000000 0.000000 0.000000
84 85 52 1 2 120.000000 325.000000 False 0.0 172.000000 False 0.200000 1.000000 0.000000 0.000000
In [432]:
y_train.head()
Out[432]:
880    1
457    0
797    1
25     0
84     0
Name: num, dtype: int64

kNN Classifier¶

In [433]:
#Fitting Classifier to the Training set
from sklearn.neighbors import KNeighborsClassifier
clf = KNeighborsClassifier()
clf.fit(x_train,y_train)
Out[433]:
KNeighborsClassifier()
In [434]:
#Predict on test data
y_pred=clf.predict(x_test)
In [435]:
print("Predicted Heart disease : ")
print(y_pred)

Predicted Heart disease : 
[0 0 1 1 1 0 1 1 0 0 0 1 1 1 1 0 0 0 1 1 0 1 1 1 1 1 1 0 1 0 0 0 0 1 0 1 0
 1 1 1 1 0 1 1 0 1 1 0 0 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 1 0 0 1 0 0 0
 0 0 1 0 1 1 1 1 0 1 1 1 1 1 0 1 1 1 0 0 0 0 0 1 0 0 0 1 1 1 1 0 1 1 0 0 1
 0 0 0 1 0 1 0 1 0 0 1 0 1 1 1 1 0 1 1 0 0 0 0 0 0 0 1 1 1 0 1 1 1 1 1 1 0
 0 1 1 1 0 1 0 1 0 0 1 0 1 1 0 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 0 1 0 1]
In [436]:
print("Actual Heart disease : ")
print(y_test.values)

Actual Heart disease : 
[0 0 1 1 1 0 1 1 1 0 1 0 1 1 1 0 0 0 0 1 0 1 1 1 1 1 1 0 1 0 1 1 0 1 0 1 0
 1 1 1 1 0 0 0 0 1 1 1 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 0
 1 0 1 0 1 0 1 1 0 0 1 1 1 1 1 1 1 1 0 1 0 1 0 1 0 0 0 0 1 1 0 0 1 1 1 0 1
 1 1 0 1 0 0 0 1 0 0 1 0 1 1 0 1 0 1 1 0 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1
 0 1 1 1 1 1 0 1 0 1 1 0 1 1 1 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 1 0 1]
In [437]:
from sklearn.metrics import accuracy_score, recall_score, roc_auc_score, confusion_matrix

print("\nAccuracy score: %f" %(accuracy_score(y_test,y_pred) * 100))

Accuracy score: 80.978261
In [438]:
print(confusion_matrix(y_test, y_pred)) 

[[63 12]
 [23 86]]
In [439]:
import matplotlib.pyplot as plt  
from sklearn.metrics import roc_curve, auc
fpr, tpr, thresholds = roc_curve(y_pred, y_test)
roc_auc = auc(fpr, tpr)

#plt.figure()
plt.plot(fpr, tpr, color='darkorange', 
         label='ROC curve (area = %0.3f)' % roc_auc)
plt.plot([0, 1], [0, 1], color='navy',  linestyle='--')
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('False Positive Rate')
plt.ylabel('True Positive Rate')
plt.title('Receiver operating characteristic')
plt.legend(loc="lower right")
plt.show()
In [440]:
print("ROC score : %f\n" %(roc_auc_score(y_test, y_pred) * 100))

ROC score : 81.449541

Random Forest Classifier¶

In [441]:
from sklearn.ensemble import RandomForestClassifier
rmf = RandomForestClassifier(max_depth=3, random_state=42)
rmf_clf = rmf.fit(x_train, y_train)
In [442]:
#Predict on test data
y_pred=rmf.predict(x_test)
In [443]:
print("Predicted Heart disease : ")
print(y_pred)

Predicted Heart disease : 
[0 0 1 1 1 0 1 1 1 0 1 1 1 1 1 0 0 0 0 1 0 0 1 1 1 1 1 0 1 0 0 0 0 1 0 1 0
 1 1 1 1 0 1 0 0 1 1 1 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 1 1 1 1 0 1 0
 0 0 0 1 0 0 1 1 0 1 1 1 1 1 1 1 1 1 0 0 0 1 0 1 0 0 1 1 1 1 0 0 1 1 1 0 0
 1 1 0 1 0 1 0 1 0 0 1 0 1 1 1 1 0 1 1 0 0 0 0 1 0 0 1 1 1 1 1 1 1 1 1 1 1
 0 1 1 1 1 1 0 1 0 1 1 0 1 1 0 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 1 0 1]
In [444]:
print("Actual Heart disease : ")
print(y_test.values)

Actual Heart disease : 
[0 0 1 1 1 0 1 1 1 0 1 0 1 1 1 0 0 0 0 1 0 1 1 1 1 1 1 0 1 0 1 1 0 1 0 1 0
 1 1 1 1 0 0 0 0 1 1 1 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 0
 1 0 1 0 1 0 1 1 0 0 1 1 1 1 1 1 1 1 0 1 0 1 0 1 0 0 0 0 1 1 0 0 1 1 1 0 1
 1 1 0 1 0 0 0 1 0 0 1 0 1 1 0 1 0 1 1 0 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1
 0 1 1 1 1 1 0 1 0 1 1 0 1 1 1 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 1 0 1]
In [445]:
print("\nAccuracy score: %f" %(accuracy_score(y_test,y_pred) * 100))
print("Recall score : %f" %(recall_score(y_test, y_pred) * 100))
print("ROC score : %f\n" %(roc_auc_score(y_test, y_pred) * 100))
print(confusion_matrix(y_test, y_pred)) 

Accuracy score: 88.586957
Recall score : 90.825688
ROC score : 88.079511

[[64 11]
 [10 99]]
In [446]:
plt.figure(dpi=150)
plt.hist(probas, bins=20)
plt.title('Classification Probabilities')
plt.xlabel('Probability')
plt.ylabel('# of Instances')
plt.xlim([0.5, 1.0])
plt.legend(y_test)
plt.show()

Navies Bayes Classification¶

In [447]:
from sklearn.naive_bayes import GaussianNB
nb_classifier =GaussianNB()
nb_classifier.fit(x_train, y_train)
Out[447]:
GaussianNB()
In [448]:
y_pred=nb_classifier.predict(x_test)
In [449]:
print("Predicted Heart disease : ")
print(y_pred)

Predicted Heart disease : 
[0 0 1 1 0 0 0 1 1 0 1 1 1 1 1 0 0 0 1 1 0 0 1 1 1 1 1 0 0 0 0 0 0 1 0 1 0
 1 0 1 1 0 1 0 0 1 1 1 0 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 0 1 0 1 1 1 1 0 1 0
 0 0 0 1 0 0 0 1 0 1 1 1 1 1 0 1 1 1 0 1 0 1 0 1 0 0 0 0 1 1 0 1 1 0 1 0 0
 1 1 0 1 0 1 0 1 0 0 1 0 1 1 0 0 0 1 0 0 0 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 1
 0 1 1 1 1 0 0 0 0 1 1 0 1 1 0 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 0 1 0 1]
In [450]:
print("Actual Heart disease : ")
print(y_test.values)

Actual Heart disease : 
[0 0 1 1 1 0 1 1 1 0 1 0 1 1 1 0 0 0 0 1 0 1 1 1 1 1 1 0 1 0 1 1 0 1 0 1 0
 1 1 1 1 0 0 0 0 1 1 1 1 0 1 0 0 1 0 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 0
 1 0 1 0 1 0 1 1 0 0 1 1 1 1 1 1 1 1 0 1 0 1 0 1 0 0 0 0 1 1 0 0 1 1 1 0 1
 1 1 0 1 0 0 0 1 0 0 1 0 1 1 0 1 0 1 1 0 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1
 0 1 1 1 1 1 0 1 0 1 1 0 1 1 1 0 1 1 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 1 0 1]
In [451]:
print("\nAccuracy score: %f" %(accuracy_score(y_test,y_pred) * 100))
print("Recall score : %f" %(recall_score(y_test, y_pred) * 100))
print("ROC score : %f\n" %(roc_auc_score(y_test, y_pred) * 100))
print(confusion_matrix(y_test, y_pred)) 

Accuracy score: 81.521739
Recall score : 78.899083
ROC score : 82.116208

[[64 11]
 [23 86]]

Pipelining¶

In [452]:
pipe_knn = Pipeline([('scl', MinMaxScaler()), 
                    ('pca', PCA(n_components=2)), 
                    ('clf', KNeighborsClassifier())])
In [453]:
pipe_rf = Pipeline([('scl', MinMaxScaler()), 
                    ('pca', PCA(n_components=3)), 
                    ('clf', RandomForestClassifier(random_state=42))])
In [454]:
pipe_nb = Pipeline([('scl', MinMaxScaler()), 
                    ('pca', PCA(n_components=2)), 
                    ('clf', GaussianNB())])
In [455]:
pipelines = [pipe_knn, pipe_rf, pipe_nb]

pipelines[1]
Out[455]:
Pipeline(steps=[('scl', MinMaxScaler()), ('pca', PCA(n_components=3)),
                ('clf', RandomForestClassifier(random_state=42))])
In [456]:
pipe_dict = {0: 'kNN Classifier', 1: 'Random Forest', 2: 'Navies Bayes'}
In [457]:
for pipe in pipelines:
	pipe.fit(x_train, y_train)
In [458]:
for idx, val in enumerate(pipelines):
	print('%s pipeline test accuracy: %.2f' % 
          (pipe_dict[idx], val.score(x_test, y_test)))

kNN Classifier pipeline test accuracy: 0.82
Random Forest pipeline test accuracy: 0.85
Navies Bayes pipeline test accuracy: 0.80
In [459]:
best_acc = 0.0
best_clf = 0
best_pipe = ''

for idx, val in enumerate(pipelines):
	if val.score(x_test, y_test) > best_acc:
		best_acc = val.score(x_test, y_test)
		best_pipe = val
		best_clf = idx

print('Classifier with best accuracy: %s' % pipe_dict[best_clf])

Classifier with best accuracy: Random Forest
In [ ]:
In [ ]: