import streamlit as st
from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline
from keybert import KeyBERT
import numpy as np
import torch
from evaluate import load

# Load sentiment analysis model and word divider
try:
    model = AutoModelForSequenceClassification.from_pretrained(
        "tabularisai/multilingual-sentiment-analysis",
        num_labels=5
    )
    tokenizer = AutoTokenizer.from_pretrained(
        "tabularisai/multilingual-sentiment-analysis"
    )
    sentiment_pipeline = pipeline(
        "text-classification",
        model=model,
        tokenizer=tokenizer
    )
except Exception as e:
    st.error(f"Error loading sentiment analysis model: {e}")

# Load keyword extraction model
try:
    kw_model = KeyBERT()
except Exception as e:
    st.error(f"Error loading sentiment analysis model: {e}")

# Define tag mapping
label_map = {
    "Very Negative": 0,
    "Negative": 1,
    "Neutral": 2,
    "Positive": 3,
    "Very Positive": 4
}

# Streamlit 
st.title("Text emotion and keyword analysis")

# Create a text entry box
input_text = st.text_area("Please enter user feedback", "")

if input_text:
    try:
        # sentiment analysis
        result = sentiment_pipeline(input_text)
        predicted_label = label_map[result[0]['label']]
        rating = predicted_label + 1  
        confidence = result[0]['score']

        # ketwords extraction
        keywords = kw_model.extract_keywords(
            input_text,
            keyphrase_ngram_range=(1, 2),
            stop_words="english",
            top_n=3
        )
        keyword_text = [kw[0] for kw in keywords]

        # result showing
        st.write(f"Rating: {rating}/5")
        st.write(f"Confidence: {confidence:.2f}")
        st.write(f"Key words: {', '.join(keyword_text)}")
    except Exception as e:
        st.error(f"Error while analyzing text: {e}")