Deep Learning with R Cookbook

Getting ready

The Boston Housing Prices dataset can be accessed directly from the keras library. It has 404 training examples and 102 test examples.

Let's load the required libraries:

library(mxnet)library(keras)

We load the dataset and split it into training and testing sets:

boston = dataset_boston_housing()train_x = boston$train$xtrain_y = boston$train$ytest_x = boston$test$xtest_y = boston$test$y

Let's now scale the dataset:

# normalize train settrain_x <- scale(train_x)# normalize test settrain_means <- attr(train_x, "scaled:center") train_stddevs <- attr(train_x, "scaled:scale")test_x <- scale(test_x, center = train_means, scale = train_stddevs)

This completes the data preprocessing part.