Deep Learning with R Cookbook

Getting ready

In this section, we will use the IMDb dataset, which contains movie reviews and sentiment associated with it. We can import dataset from the keras library. These reviews are preprocessed and encoded as a sequence of word indexes. These words are indexed by their overall frequency in the dataset; for example, the word index 8, refers to the 8^th most frequent word in the data.

Now, let's import the keras library and the imdb dataset:

library(keras)imdb <- dataset_imdb(num_words = 1000)

Let's divide the data into training and testing sets:

train_x <- imdb$train$xtrain_y <- imdb$train$ytest_x <- imdb$test$xtest_y <- imdb$test$y

Now, we can have a look at the number of reviews in the train and test data:

# number of samples in train ...