# import part
import streamlit as st
from transformers import pipeline, AutoTokenizer
import numpy as np
import random
import torch

# Set random seed for reproducibility
seed = 42
random.seed(seed)
np.random.seed(seed)
torch.manual_seed(seed)
if torch.cuda.is_available():
    torch.cuda.manual_seed_all(seed)

# Load the tokenizer for the sentiment analysis model
tokenizer = AutoTokenizer.from_pretrained("nlptown/bert-base-multilingual-uncased-sentiment")

# Initialize sentiment analysis pipeline with the custom model and tokenizer
sentiment_pipeline = pipeline("sentiment-analysis", model="EmmaL1/CustomModel_amazon", tokenizer=tokenizer)

# Initialize question-answering pipeline
qa_pipeline = pipeline("question-answering", model="distilbert/distilbert-base-cased-distilled-squad")

# function part
# text classification
def textclassification():
    st.title("Amazon Customer Sentiment Analysis: Ratings and Reasons")
    st.write("Enter a sentence to analyze its rating and reason:")

    user_input = st.text_input("Input your text:")
    if user_input:
        # Sentiment Analysis
        sentiment_result = sentiment_pipeline(user_input)
        sentiment = sentiment_result[0]["label"]
        confidence = sentiment_result[0]["score"]

        st.write(f"Sentiment: {sentiment}")
        st.write(f"Confidence: {confidence:.2f}")

        # Display the rating
        st.write(f"The rating is {sentiment}")

        # Question Answering
        qa_input = {
            'question': f'Why is the rating {sentiment} star?',
            'context': user_input  # Use user input as context
        }
        qa_result = qa_pipeline(qa_input)
        st.write(f"Reasons: {qa_result['answer']}")   

def main():
    textclassification()

if __name__ == "__main__":
    main()