import streamlit as st
import tensorflow as tf
import pickle
import numpy as np
from tensorflow.keras.preprocessing.sequence import pad_sequences

# Load the model and tokenizer
model = tf.keras.models.load_model('sentiment_model.keras')

with open('tokenizer.pickle', 'rb') as handle:
    tokenizer = pickle.load(handle)

with open('max_length.txt', 'r') as f:
    max_length = int(f.read())

def classify_sentence(sentence):
    seq = tokenizer.texts_to_sequences([sentence])
    padded_seq = pad_sequences(seq, maxlen=max_length)
    prediction = model.predict(padded_seq)
    label = "Positive" if prediction[0][0] > 0.5 else "Negative"
    return label

# Streamlit UI
st.title("Restaurant Review Sentiment Analysis")
st.write("Enter your review in Turkish to analyze its sentiment")

user_input = st.text_area("Enter your review:")

if st.button("Analyze"):
    if user_input:
        result = classify_sentence(user_input)
        st.write(f"Sentiment: {result}")
    else:
        st.write("Please enter a review to analyze")