import streamlit as st
from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch
# Streamlit UI
st.set_page_config(page_title="Emotion Detection", layout="centered")
# 🎯 Add Image at the Top
st.image("inno.jpg", use_container_width=True,width=300) 
#1. Business and Data Understanding
st.header(" 1. Business and Data Understanding")
st.write("""
         Understanding human emotions is critical for businesses that interact with customers, employees, or users through text, speech, or images.
         Subjectivity of Emotions: Emotions can be complex, overlapping, and vary based on individual perceptions.
* Data Limitations: Labeled emotional datasets are often limited or imbalanced.
* Real-time Processing Needs: Businesses require fast, scalable solutions for handling large volumes of data.
* Multimodal Complexity: Emotions can be expressed through text, speech, or facial expressions, requiring different ML models.
* Privacy & Ethical Concerns: Emotion analysis can raise ethical and legal issues related to data privacy.""")
# b) Business Objective
st.subheader(" b) Business Objective")
st.write("""
The primary business objective of emotion detection is to understand, analyze, and
          respond to human emotions in real time to improve customer experience,
          employee engagement, marketing strategies, and decision-making. 
Key Business Objectives:
1 Improve Customer Experience & Satisfaction:
- Detect customer emotions in support chats, emails, and voice calls.
- Provide personalized responses and proactive service to reduce customer frustration.
- Optimize chatbots and virtual assistants to improve engagement.

2 Enhance Brand Reputation & Market Intelligence:
- Analyze social media sentiments and product reviews to measure public perception.
- Identify negative feedback early to prevent PR crises.
- Improve product and service offerings based on emotional insights.

3 Increase Employee Engagement & Workplace Well-being:
-  Analyze employee sentiment in feedback surveys, emails, and communication tools.
- Detect burnout, dissatisfaction, or stress levels in employees.
- Improve HR policies and work culture based on emotion analytics.

4 Optimize Marketing & Advertising Strategies:
- Measure emotional reactions to ads, videos, and brand campaigns.
- Personalize marketing content based on user emotions.
- Improve product recommendations and customer targeting.

5 Support Mental Health & Well-being:
- Detect signs of stress, anxiety, or depression in conversations.
-  Provide AI-driven emotional support through chatbots and virtual therapy.
- Assist psychologists and therapists in tracking patient emotions.

6 Enable Real-Time Decision Making:
- Use emotion-based insights to make faster, data-driven business decisions.
-  Improve customer retention by addressing negative sentiments proactively.
- Automate sentiment monitoring for large-scale data analysis.""")
# c) Business Constraints
st.subheader(" c) Business Constraints")
st.write("""
- *Data Privacy & Security*
- *Language & Emoji Variations*
- *Accuracy & Context Awareness*
- *scalability*
- *Real-Time Processing*
""")

# d) Data Understanding
st.subheader(" d) Data Understanding")
st.write("""
- *Dataset Size:* 422,746 text samples.
- *Data Types:*  Text data, labeled emotions, unstructured text.
- *Data Features:*
 *Raw Text:* Actual emotional expressions.
 *Emojis & Symbols:* Indicators of sentiment.
""")


# Load pre-trained emotion detection model
model_name = "j-hartmann/emotion-english-distilroberta-base"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Get the actual emotion labels from the model
emotion_labels = model.config.id2label

# Emotion styles (emoji + colors)
emotion_styles = {
    "joy": {"emoji": "😃", "color": "#D3D3D3"},        
    "sadness": {"emoji": "😢", "color": "#3498BB"},    
    "anger": {"emoji": "😡", "color": "#FFDAA9"},      
    "fear": {"emoji": "😨", "color": "#FFFAAD"},       
    "surprise": {"emoji": "😲", "color": "#98CB98"},   
    "disgust": {"emoji": "🤢", "color": "#FFBFC1"},          
    "neutral": {"emoji": "😐", "color": "#E6E6FA"}    
}


# 🎨 Change Background Color
st.markdown(
    """
    <style>
    body {
        background-color:#AFEEEE; /* Light Grayish Blue */
    }
    </style>
    """,
    unsafe_allow_html=True
)



st.markdown("<h1 style='text-align: center; color: #7B3B98;'>Emotion Detection using ML</h1>", unsafe_allow_html=True)
st.markdown("<h3 style='text-align: left; color: #3498BB;'>📝 Enter the text:</h3>", unsafe_allow_html=True)

# User Input
user_text = st.text_input("", placeholder="Type your text here...")

if st.button("Submit"):
    if user_text:
        # Tokenize input text
        inputs = tokenizer(user_text, return_tensors="pt")

        # Get model predictions
        with torch.no_grad():


            
            outputs = model(**inputs)

        # Get the predicted emotion
        scores = outputs.logits[0]
        predicted_label_id = torch.argmax(scores).item()
        predicted_emotion = emotion_labels[predicted_label_id].strip().lower()

        # Get emoji & color
        emotion_data = emotion_styles.get(predicted_emotion, {"emoji": "😐", "color": "#D3D3D3"})
        emoji_display = emotion_data["emoji"]
        text_color = emotion_data["color"]

        # Display Results with Color
        st.markdown(
            f"""
            <div style="text-align: center; padding: 10px; border-radius: 10px; background-color: {text_color}; color: black; font-size: 24px;">
                <b>Detected Emotion:</b> {predicted_emotion.capitalize()} {emoji_display}
            </div>
            """,
            unsafe_allow_html=True
        )
    else:
        st.warning("Please enter some text!")