Code.py

Implementation of the all the supervised learning algorithms on the breast cancer datasets.

Author: iGyandeep

Code.py

@@ -0,0 +1,102 @@
+‘‘‘Importing the function’’’
+from sklearn import tree
+from sklearn.neural_network import MLPClassifier
+#import matplotlib.pyplot as plt
+from sklearn.naive_bayes import GaussianNB as p
+from sklearn.metrics import confusion_matrix
+from sklearn.neighbors import KNeighborsClassifier
+from sklearn.model_selection import cross_val_score
+from sklearn import svm
+import math
+import numpy as np
+import pandas as pd
+
+#‘‘‘K-Nearest Neighbors Function’’’
+
+
+def KNN(x_training,y_training,x_testing,y_testing):
+  print("\nKNN")
+  clf=KNeighborsClassifier(n_neighbors=9)
+  clf.fit(x_training,y_training)
+  cvs=cross_val_score(clf,x_testing,y_testing,cv=5)
+  result=clf.predict(x_testing)
+  cmknn=confusion_matrix(y_testing,result)
+  print("Confusion_Matrix\n\n",cmknn)
+  print("Accuracy =",((cmknn[0][0]+cmknn[1][1])/len(y_testing))*100)
+  print("Mean=",cvs.mean())
+
+#‘‘‘NaiveB Function’’’
+
+def NaiveB(x_training,y_training,x_testing,y_testing):
+  print("\n\nNaiveBaysian")
+  gnb=p()
+  gnb.fit(x_training,y_training)
+  y_pre=gnb.predict(x_testing)
+  cm=confusion_matrix(y_testing,y_pre)
+  print("Confusion_Matrix\n",cm)
+  print("Accuracy=",((cm[0][0]+cm[1][1])/len(y_testing)*100))
+
+#‘‘‘Decision Tree Function’’’
+
+def DecisionTree(x_training,y_training,x_testing,y_testing):
+  print("\n\nDecision Tree")
+  clf=tree.DecisionTreeClassifier()
+  clf=clf.fit(x_training,y_training)
+  y_predict=clf.predict(x_testing)
+  CM=confusion_matrix(y_testing,y_predict)
+  accuracy=sum(CM.diagonal())/len(y_test)
+  print("Confusion_Matrix\n",CM)
+  print("Accuracy=",accuracy*100)
+
+#‘‘‘Artificial Neural Network Function’’’
+
+def NeuralNetwork(x_training,y_training,x_testing,y_testing):
+  print("\n\n Artificial Neural Network")
+  
+  clf = MLPClassifier(solver='lbfgs', alpha=1e-
+  5,hidden_layer_sizes=(5,3),random_state=1) #default hidden layer=100
+  
+  clf.fit(x_training,y_training) #The default solver ‘adam’ works pretty well on
+  relatively large datasets
+  y_predict=clf.predict(x_testing)
+  cm=confusion_matrix(y_testing,y_predict)
+  print("Confusion matrix\n\n",cm)
+  print("Accuracy=",sum(cm.diagonal())/len(x_testing)*100)
+
+#‘‘‘Support Vector Machine Function’’’
+
+def SVM(x_training,y_training,x_testing,y_testing):
+  print("\n\nSVM")
+  clf = svm.SVC(gamma='auto')
+  clf.fit(x_training,y_training)
+  y_predict=clf.predict(x_testing)
+  cm=confusion_matrix(y_testing,y_predict)
+  print("Confusion matrix\n\n",cm)
+  print("Accuracy=",sum(cm.diagonal())/len(x_testing)*100)
+
+#‘‘‘Algorithm’’’
+
+def algorithm(x_train,y_train,x_test,y_test):
+  acc=[]
+  NaiveB(x_train,y_train,x_test,y_test,acc)
+  KNN(x_train,y_train,x_test,y_test,acc)
+  DecisionTree(x_train,y_train,x_test,y_test,acc)
+  NeuralNetwork(x_train,y_train,x_test,y_test,acc)
+  SVM(x_train,y_train,x_test,y_test,acc)
+
+#‘‘‘Importing csv file’’’
+
+d=pd.read_csv("breast_cancer_weka_dataset.csv")
+c=d.copy()
+x_train=c.loc[:350,:"mitosis"]
+y_train=c.loc[:350,"class"]
+x_test=c.loc[350:,:"mitosis"]
+y_test=c.loc[350:,"class"]
+
+#‘‘‘Applying function’’’
+
+NaiveB(x_train,y_train,x_test,y_test)
+KNN(x_train,y_train,x_test,y_test)
+DecisionTree(x_train,y_train,x_test,y_test)
+NeuralNetwork(x_train,y_train,x_test,y_test)
+SVM(x_train,y_train,x_test,y_test)