prep.make_dictionary()
# Encode all words with integer IDs
# Encode only the most used words in the dataset, any other words encode as 0
n_top_used_words = 10000
dataset = prep.encode_dataset_column(df=dataset, field="review", use_top_words=n_top_used_words)
# Encode target variables to binary representation
dataset = prep.string_to_int(df=dataset, params={"sentiment": {'positive': 1, 'negative': 0}})
# Pad all reviews, remove reviews that have no words, trim reviews that exceed the review_len value
review_len = 500
dataset = prep.pad_text(df=dataset, column="review_encoded", min_words=1, max_words=review_len)
# Split the dataset into training, test and validation subsets
train_s, test_s, valid_s = prep.split_dataset(training_r=0.5, test_r=0.3, validation_r=0.2, dataset=dataset)
# Convert dataframe column to the numpy array
X_train = np.array(train_s['review_encoded'].tolist())
Y = np.array(train_s['sentiment'].tolist())
X_eval = np.array(valid_s['review_encoded'].tolist())
Yv = np.array(valid_s['sentiment'].tolist())
X_test = np.array(test_s['review_encoded'].tolist())
Yt = np.array(test_s['sentiment'].tolist())
# ************************************************** #
# THE SIMPLE RNN MODEL #
# ************************************************** #
dataset = prep.string_to_int(
df=dataset, params={"sentiment": {
'positive': 1,
'negative': 0
}})
# Pad all reviews, remove reviews that have no words, trim reviews that exceed the review_len value
review_len = 500
dataset = prep.pad_text(df=dataset,
column="review_encoded",
min_words=1,
max_words=review_len)
# Split the dataset into training, test and validation subsets
train_s, test_s, valid_s = prep.split_dataset(training_r=0.5,
test_r=0.3,
validation_r=0.2,
dataset=dataset)
# Convert dataframe column to the numpy array
X_train = np.array(train_s['review_encoded'].tolist())
Y = np.array(train_s['sentiment'].tolist())
X_eval = np.array(valid_s['review_encoded'].tolist())
Yv = np.array(valid_s['sentiment'].tolist())
X_test = np.array(test_s['review_encoded'].tolist())
Yt = np.array(test_s['sentiment'].tolist())
# ************************************************** #
# MODELS COMMON SETTINGS #
# ************************************************** #