build a simple neural network in r

main.r
# Load the neuralnet package
library(neuralnet)

# Load the iris dataset
data(iris)

# Split data into training and testing sets
train <- iris[1:100, ]
test <- iris[101:150, ]

# Preprocess data by scaling the input variables
train[, 1:4] <- scale(train[, 1:4])
test[, 1:4] <- scale(test[, 1:4])
296 chars
14 lines

main.r
# Build the neural network model
nn <- neuralnet(Species ~ Sepal.Length + Sepal.Width + Petal.Length + Petal.Width, 
                data = train, hidden = 5)
159 chars
4 lines

main.r
# Plot the neural network
plot(nn)
35 chars
3 lines

main.r
# Make predictions on the testing data
predictions <- compute(nn, test[, 1:4])

# Convert the output to a factor
predicted.species <- factor(max.col(predictions$net.result), 
                            levels = 1:3, labels = levels(iris$Species))

# Check the accuracy of the predictions
mean(predicted.species == test$Species) 
330 chars
10 lines