Padding the text sequences is carried out to ensure that all the sequences have the same length. Let's take a look at the following code:

# Padding and truncation
train_x <- pad_sequences(train_x, maxlen = 100)
test_x <- pad_sequences(test_x, maxlen = 100)

From the preceding code, we can observe the following:

• We can achieve equal length for all the sequences of integers with the help of the pad_sequences function and by specifying a value for maxlen.
• In this example, we have restricted the length of each movie review sequence in the train and test data to 100. Note that before padding of sequences, the structure of train_x and test_x is a list of 25,000 reviews.
• However, after padding the sequences, the structure for both changes to a matrix that's 25,000 x 100. This can be easily verified by running str(train_x) before and after padding.

To observe the impact of padding on a sequence of integers, let's take a look at the following code, along with its output:

# Sequence of integers
train_x[3,]
  [1]   2   4   2  33  89  78  12  66  16   4 360   7   4  58 316 334
 [17]  11   4   2  43   2   2   8 257  85   2  42   2   2  83  68   2
 [33]  15  36 165   2 278  36  69   2   2   8 106  14   2   2  18   6
 [49]  22  12 215  28   2  40   6  87 326  23   2  21  23  22  12 272
 [65]  40  57  31  11   4  22  47   6   2  51   9 170  23   2 116   2
 [81]   2  13 191  79   2  89   2  14   9   8 106   2   2  35   2   6
 [97] 227   7 129 113

train_x[6,]
  [1]   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
 [17]   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
 [33]   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
 [49]   0   0   0   0   0   0   0   0   0   1   2 128  74  12   2 163
 [65]  15   4   2   2   2   2  32  85 156  45  40 148 139 121   2   2
 [81]  10  10   2 173   4   2   2  16   2   8   4 226  65  12  43 127
 [97]  24   2  10  10

The output of the third sequence of integers after padding of the train_x can be seen in the preceding code. Here, we can observe the following:

• The third sequence now has a length of 100. The third sequence originally had 141 integers and we can observe that 41 integers that were located at the beginning of the sequence have been truncated.
• On the other hand, the output of the sixth sequence shows a different pattern.
• The sixth sequence originally had a length of 43, but now 57 zeros have been added to the beginning of the sequence to artificially extended the length to 100.
• All 25,000 sequences of integers related to movie reviews in each of the train and test data are impacted in a similar way.

In the next section, we will develop an architecture for a recurrent neural network that will be used for developing a movie review sentiment classification model.