import pandas as pd
import tensorflow as tf

train_data = pd.read_csv('./beer_reviews_train.csv')
X_train = train_data[['review_aroma', 'review_appearance', 'review_palate', 'review_taste']]
y_train = train_data['review_overall']

tokenizer = tf.keras.preprocessing.text.Tokenizer(num_words=10000)
tokenizer.fit_on_texts(X_train)
X_train_seq = tokenizer.texts_to_sequences(X_train)

X_train_pad = tf.keras.preprocessing.sequence.pad_sequences(X_train_seq, maxlen=100)

model = tf.keras.Sequential([
    tf.keras.layers.Embedding(input_dim=10000, output_dim=16, input_length=100),
    tf.keras.layers.GlobalAveragePooling1D(),
    tf.keras.layers.Dense(16, activation='relu'),
    tf.keras.layers.Dense(1, activation='sigmoid')
])

model.compile(optimizer='adam',
              loss='binary_crossentropy',
              metrics=['accuracy'])

model.fit(X_train_pad, y_train, epochs=40, batch_size=32, validation_split=0.1)

model.save('beer_review_sentiment_model.h5')