Machine Learning Model

#!/usr/bin/env python
# coding: utf-8

# In[16]:


import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import glob
import os

import plotly.express as px
import plotly.offline as py
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC
from sklearn.metrics import confusion_matrix
from sklearn.metrics import accuracy_score
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import precision_score
from sklearn.metrics import recall_score
from sklearn.metrics import roc_curve
from sklearn.metrics import roc_auc_score
from sklearn.metrics import auc
from sklearn.metrics import precision_recall_curve
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import roc_auc_score
import matplotlib.pyplot as plt
from sklearn.metrics import accuracy_score
from sklearn import svm


# In[17]:


df=pd.read_csv('D:/heart.csv')
df


# In[18]:


# 1.    age
# 2.    sex (1 = male; 0= female)
# 3.    chest pain type (4 values)
# 4.    resting blood pressure
# 5.    serum cholestoral in mg/dl
# 6.    fasting blood sugar > 120 mg/dl
# 7.    resting electrocardiographic results (values 0,1,2)
# 8.    maximum heart rate achieved
# 9.    exercise induced angina
# 10.   oldpeak = ST depression induced by exercise relative to rest
# 11.   the slope of the peak exercise ST segment
# 12.   number of major vessels (0-3) colored by flourosopy
# 13.   thal: 3 = normal; 6 = fixed defect; 7 = reversable defect

df.describe()


# In[19]:


df.head()


# In[21]:


# Gender Distribution
df['sex']=df['sex'].map({
    1:'Male',
    0:'Female'
})
px.pie(df, names='sex',
#        color_discrete_sequence=px.colors.sequential.Aggrnyl,
#        hole=.6,

       )


# In[123]:


df.isnull().sum()


# In[124]:


plt.figure(figsize=(20,10))
sns.countplot(x=df['age'],data=df)
plt.xticks(rotation=90,fontsize=30)
plt.yticks(fontsize=30)


# In[24]:


px.histogram(df,
             x='age',
             hover_data=df.columns,
             marginal="box",
             color='target')


# In[27]:


px.violin(df, y="chol",
           color="target",
           hover_data=df.columns,
            points="all",
           box=True,
          )


# In[28]:


px.violin(df, y="age",
           color="target",
           hover_data=df.columns,
            points="all",
           box=True,
          )


# In[29]:


sns.pairplot(df)
plt.show()


# In[30]:


plt.figure(figsize=(15,15))
sns.heatmap(df.corr()) #corelation
plt.show()


# In[31]:


df.corr()


# In[32]:



pred=df.drop(["target","fbs","chol","age","sex","trestbps"],axis=1)
target=df['target']
X_train,X_test,Y_train,Y_test = train_test_split(pred,target,test_size=0.20)


# In[33]:


X_train.shape


# In[34]:


X_test.shape


# In[35]:


def training_model(X_train, y_train, X_test, y_test, classifier, **kwargs):
    model=classifier(**kwargs)
    model.fit(X_train,y_train)
    train_accuracy=model.score(X_train,y_train)
    test_accuracy=model.score(X_test,y_test)
    print(f"Train accuracy: {train_accuracy:0.3%}")
    print(f"Test accuracy: {test_accuracy:0.3%}")

    return model


# In[36]:



KNN=training_model(X_train, Y_train, X_test, Y_test, KNeighborsClassifier, n_neighbors=7)
KNN.fit(X_train,Y_train)
KNN.score(X_train,Y_train)
y_pred_test=KNN.predict(X_test)
y_pred_train=KNN.predict(X_train)
# print (accuracy_score(Y_test, y_pred_test))


# In[37]:



matrix=confusion_matrix(Y_test,y_pred_test)
sns.heatmap(matrix,annot=True)


# In[38]:



precision = precision_score(Y_test, y_pred_test)
recall = recall_score(Y_test, y_pred_test)
print("Precision: ",precision)
print("Recall is: ",recall)
print("fbeta score:",(2*precision*recall)/(precision+recall))


# In[41]:



y_pred_prob = KNN.predict_proba(X_test)[:,1]
FPR, TPR, threshold = roc_curve(Y_test, y_pred_prob)



# In[55]:



SVM = svm.SVC(kernel='linear',C=5,probability=True)
SVM.fit(X_train, Y_train)
yhat1_test = SVM.predict(X_test)
yhat1_train=SVM.predict(X_train)
print ("test_acc_score",accuracy_score(Y_test, yhat1_test))
print ("train_acc_score",accuracy_score(Y_train, yhat1_train))


# In[48]:



matrix1=confusion_matrix(Y_test,yhat1_test)
sns.heatmap(matrix1,annot=True)


# In[59]:


precision = precision_score(Y_test,yhat1_test)
recall = recall_score(Y_test, yhat1_test)
print("Precision: ",precision)
print("Recall is: ",recall)
print("fbeta score is:",(2*precision*recall)/(precision+recall))


# In[56]:


y_pred_prob1 = SVM.predict_proba(X_test)[:,1]
FPR1, TPR1, threshold1 = roc_curve(Y_test, y_pred_prob1)


# In[57]:


# matplotlib

plt.style.use('seaborn')

# plot roc curves
plt.plot(FPR, TPR, linestyle='--',color='green', label='KNN')
plt.plot(FPR1, TPR1, linestyle='--',color='red', label='SVM')

# title
plt.title('ROC curve')
# x label
plt.xlabel('False Positive Rate')
# y label
plt.ylabel('True Positive rate')

plt.legend(loc='best')
plt.savefig('ROC',dpi=300)
plt.show();


# In[58]:



AUC_score1 = roc_auc_score(Y_test, y_pred_prob)
AUC_score2 = roc_auc_score(Y_test, y_pred_prob1)

print(AUC_score1)
print(AUC_score2)


# In[ ]:


# parameter tuning for SVM using gridsearch
svm_clf = SVC(kernel='rbf', gamma=0.1, C=1.0)

params = {"C":(0.1, 0.5, 1, 2, 5, 10, 20),
          "gamma":(0.001, 0.01, 0.1, 0.25, 0.5, 0.75, 1),
          "kernel":('linear', 'poly', 'rbf')}

svm_cv = GridSearchCV(svm_clf, params, n_jobs=-1, cv=5, verbose=1, scoring="accuracy")
svm_cv.fit(X_train, Y_train)
best_params = svm_cv.best_params_
print(f"Best params: {best_params}")

svm_clf = SVC(**best_params)
svm_clf.fit(X_train, Y_train)

print_score(svm_clf, X_train, Y_train, X_test, Y_test, train=True)
print_score(svm_clf, X_train, Y_train, X_test, Y_test, train=False)