import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics import accuracy_score
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Embedding, LSTM
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.utils import to_categorical

# Step 1: Data Preprocessing
df = pd.read_csv('25k_movies.csv.shuf')  
text_data = df['review']
labels = df['sentiment']

# Step 2: Data Split
X_train, X_test, y_train, y_test = train_test_split(text_data, labels, test_size=0.2, random_state=42)
X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size=0.2, random_state=42)

# Step 3: Vectorization
vectorizer = TfidfVectorizer()
X_train_vec = vectorizer.fit_transform(X_train)
X_val_vec = vectorizer.transform(X_val)
X_test_vec = vectorizer.transform(X_test)

# Step 4: Model Architecture
model = Sequential()
model.add(Dense(128, activation='relu', input_shape=(X_train_vec.shape[1],)))
model.add(Dense(64, activation='relu'))
model.add(Dense(1, activation='sigmoid'))

# Step 5: Training
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
model.fit(X_train_vec, y_train, batch_size=32, epochs=10, validation_data=(X_val_vec, y_val))

# Step 6: Evaluation
y_pred = model.predict_classes(X_test_vec)
accuracy = accuracy_score(y_test, y_pred)
print("Test Accuracy:", accuracy)

# Step 7: Fine-tuning and Optimization
# Adjust hyperparameters, architecture, and retrain the model as needed

# Step 8: Inference
new_reviews = ['Great movie!', 'Terrible acting.']
new_reviews_vec = vectorizer.transform(new_reviews)
predictions = model.predict_classes(new_reviews_vec)
sentiments = ['Positive' if p == 1 else 'Negative' for p in predictions]
print("Predictions:", sentiments)