import tensorflow as tf
import matplotlib.pyplot as plt
import pandas as pd
df = pd.read_csv("IMDB Dataset(2).csv")
df['sentiment'] = df['sentiment'].map({'positive':1,'negative':0})
train = df.sample(frac=0.8,random_state=42)
test = df.drop(train.index)
x_train = train['review'].astype(str).values
y_train = train['sentiment'].values
x_test = test['review'].astype(str).values
y_test = test['sentiment'].values
tokenizer = tf.keras.preprocessing.text.Tokenizer(num_words=5000)
tokenizer.fit_on_texts(x_train)
x_train = tokenizer.texts_to_sequences(x_train)
x_test = tokenizer.texts_to_sequences(x_test)
x_train = tf.keras.preprocessing.sequence.pad_sequences(x_train,maxlen=100)
x_test = tf.keras.preprocessing.sequence.pad_sequences(x_test,maxlen=100)
model = tf.keras.Sequential([
    tf.keras.layers.Embedding(5000,64),
    tf.keras.layers.SimpleRNN(32), # Changed from LSTM(64) to SimpleRNN(32)
    tf.keras.layers.Dense(1,activation='sigmoid')
])
model.compile(optimizer='adam',loss='binary_crossentropy',metrics=['accuracy'])
h = model.fit(x_train,y_train,epochs=3,batch_size=64,validation_split=0.1)
print("Test Accuracy:",model.evaluate(x_test,y_test,verbose=0)[1]*100,"%")
plt.plot(h.history['accuracy'],label='Train')
plt.plot(h.history['val_accuracy'],label='Validation')
plt.title("Simple RNN IMDB Accuracy")
plt.xlabel("Epochs")
plt.legend()
plt.show()
for i in range(3):
    p = model.predict(x_test[i:i+1],verbose=0)[0][0]
    print("Review",i+1,":","Positive" if p>=0.5 else "Negative",round(p,2))
review = ["This movie was amazing and I really enjoyed it"]
seq = tf.keras.preprocessing.sequence.pad_sequences(tokenizer.texts_to_sequences(review),maxlen=100)
p = model.predict(seq,verbose=0)[0][0]
print("Prediction:", "Positive" if p>=0.5 else "Negative")
