6. LSTMs, GRUs, and Advanced RNNs

Activity 6.01: Sentiment Analysis of Amazon Product Reviews

Solution

1. Read in the data files for the train and test sets. Examine the shapes of the datasets and print out the top 5 records from the train data:

   import pandas as pd, numpy as np
   import matplotlib.pyplot as plt
   %matplotlib inline
   train_df = pd.read_csv("Amazon_reviews_train.csv")
   test_df = pd.read_csv("Amazon_reviews_test.csv")
   print(train_df.shape, train_df.shape)
   train_df.head(5)

   The dataset's shape and header are as follows:

   Figure 6.26: First five records from the train dataset

2. For convenience, when it comes to processing, separate the raw text and the labels for the train and test sets. You should have 4 variables, as follows: train_raw comprising raw text for the train data, train_labels with labels for the train data, test_raw containing raw text for the test data, and test_labels comprising Labels for the test data. Print the first two reviews...