# Install necessary libraries
# pip install streamlit transformers datasets

import streamlit as st
from transformers import pipeline

# Load pre-trained model from Hugging Face
emotion_analyzer = pipeline("text-classification", model="distilbert-base-uncased-finetuned-sst-2")

# Define the function to analyze emotions and suggest strategies
def analyze_and_suggest(responses):
    suggestions = []
    for response in responses:
        # Get the sentiment analysis result
        result = emotion_analyzer(response)[0]
        label = result['label']

        # Suggest strategies based on sentiment
        if label == "NEGATIVE":
            suggestions.append("Try deep breathing exercises or mindfulness activities.")
        elif label == "POSITIVE":
            suggestions.append("Great! Keep the positivity going with a walk or some light exercise.")
        else:
            suggestions.append("Consider focusing on better sleep or reflecting on your priorities.")
    
    return suggestions

# Streamlit App
st.title("Personalized Self-Care Strategy App")
st.markdown("### Answer the following questions to get personalized self-care suggestions.")

# List of questions
questions = [
    "1. How do you feel about your overall health today?",
    "2. How have you been sleeping recently?",
    "3. Do you feel overwhelmed with tasks or emotions?",
    "4. What are your energy levels like today?",
    "5. How often do you exercise or engage in physical activity?"
]

# Collect user inputs
responses = []
for question in questions:
    responses.append(st.text_input(question, placeholder="Type your response here..."))

# Button to analyze and provide suggestions
if st.button("Get Self-Care Suggestions"):
    if all(responses):  # Ensure all questions are answered
        suggestions = analyze_and_suggest(responses)
        st.markdown("### **Your Personalized Suggestions**")
        for i, suggestion in enumerate(suggestions, 1):
            st.write(f"**{i}.** {suggestion}")
    else:
        st.error("Please answer all the questions before proceeding.")