library(mlbench)
library(keras)
data2<-data(BreastCancer)
data2<-BreastCancer
data2<-na.omit(data2)
data2<-data2[-1]
for(i in 1:ncol(data2)){data2[,i]<-as.numeric(as.factor(data2[,i]))}
data2<-as.matrix(data2)
data2[,10]<-as.numeric(as.factor(data2[,10]))
split<-sample(1:2,nrow(data2),prob = c(0.8,0.2),replace = T)
train<-data2[split==1,]
test<-data2[split==2,]
trainx<-train[,1:9]
train_y<-as.numeric(as.factor(train[,10]))
train_y<-train_y-1
y_train<- to_categorical(train_y)
testx<-test[,1:9]
test_y<-as.numeric(as.factor(test[,10]))
test_y<-test_y-1
y_test<-to_categorical(test_y)
model<-keras_model_sequential()
model %>%
layer_dense(units =512,activation = 'relu',input_shape = 9) %>%
layer_dense(units = 10, activation = 'relu')%>%
layer_dense(units = 10, activation = 'relu')%>%
layer_dense(units = 2, activation = 'sigmoid')
model %>% compile(
loss='binary_crossentropy',
optimizer='adam',
metrics=c('accuracy')
)
summary(model)
## Model: "sequential"
## ___________________________________________________________________________
## Layer (type) Output Shape Param #
## ===========================================================================
## dense (Dense) (None, 512) 5120
## ___________________________________________________________________________
## dense_1 (Dense) (None, 10) 5130
## ___________________________________________________________________________
## dense_2 (Dense) (None, 10) 110
## ___________________________________________________________________________
## dense_3 (Dense) (None, 2) 22
## ===========================================================================
## Total params: 10,382
## Trainable params: 10,382
## Non-trainable params: 0
## ___________________________________________________________________________
trainx<-as.matrix(trainx)
class(trainx)
## [1] "matrix"
history <- model %>% fit(trainx,y_train,epochs=50,batch_size=5,validation_split=0.2)
plot(history)
score <- model %>% evaluate(testx, y_test)
cat('Test loss:', score$loss, "\n")
## Test loss: 0.05903848
cat('Test accuracy:', score$acc, "\n")
## Test accuracy: 0.9794521
sam<-predict_classes(model,testx)
library(caret)
## Loading required package: lattice
## Loading required package: ggplot2
table(sam,test_y)
## test_y
## sam 0 1
## 0 97 2
## 1 1 46
confusionMatrix(table(sam,test_y))
## Confusion Matrix and Statistics
##
## test_y
## sam 0 1
## 0 97 2
## 1 1 46
##
## Accuracy : 0.9795
## 95% CI : (0.9411, 0.9957)
## No Information Rate : 0.6712
## P-Value [Acc > NIR] : <2e-16
##
## Kappa : 0.9532
##
## Mcnemar's Test P-Value : 1
##
## Sensitivity : 0.9898
## Specificity : 0.9583
## Pos Pred Value : 0.9798
## Neg Pred Value : 0.9787
## Prevalence : 0.6712
## Detection Rate : 0.6644
## Detection Prevalence : 0.6781
## Balanced Accuracy : 0.9741
##
## 'Positive' Class : 0
##