app.py

# import gradio as gr

# gr.load("models/dima806/facial_emotions_image_detection").launch()

import gradio as gr
import tensorflow as tf
from transformers import pipeline

def process_image(image):
    try:
        # Initialize emotion classifier
        classifier = pipeline("image-classification", model="dima806/facial_emotions_image_detection")
        
        # Get predictions
        result = classifier(image)
        
        # Convert results to required format
        emotions = {item['label']: float(item['score']) * 100 for item in result}
        
        # Generate analysis and report
        analysis = generate_analysis(emotions)
        report = generate_report(emotions, analysis)
        
        return [(k, v) for k, v in emotions.items()], analysis['detailed_analysis'], report
    except Exception as e:
        return [("error", 0)], f"Error: {str(e)}", "Error generating report"

def generate_analysis(emotions):
    descriptions = {
        'happy': 'indicates joy and positive mood',
        'angry': 'suggests frustration or displeasure',
        'surprise': 'shows astonishment',
        'fear': 'reflects anxiety or concern',
        'neutral': 'displays a balanced state'
    }
    
    top_emotion = max(emotions.items(), key=lambda x: x[1])
    
    return {
        "primary_emotion": top_emotion[0],
        "detailed_analysis": f"Primary emotion detected is {top_emotion[0]} ({top_emotion[1]:.1f}%), which {descriptions.get(top_emotion[0], '')}."
    }

def generate_report(emotions, analysis):
    return f"""Emotion Analysis Report

Primary Emotion: {analysis['primary_emotion'].upper()}
Confidence: {emotions[analysis['primary_emotion']]:.1f}%

All Detected Emotions:
""" + '\n'.join(f"- {k}: {v:.1f}%" for k, v in sorted(
        emotions.items(), 
        key=lambda x: x[1], 
        reverse=True))

# Create Gradio interface
with gr.Blocks(theme=gr.themes.Soft()) as app:
    gr.Markdown("# Facial Emotion Analysis System")
    
    with gr.Row():
        with gr.Column(scale=1):
            input_image = gr.Image(type="numpy")
            submit_btn = gr.Button("Analyze", variant="primary")
            
        with gr.Column(scale=1):
            emotion_scores = gr.Label(label="Emotion Scores")
            analysis_text = gr.Textbox(label="Analysis", lines=3)
            report_text = gr.Textbox(label="Full Report", lines=10)
            download_btn = gr.Button("Download Report")
    
    submit_btn.click(
        fn=process_image,
        inputs=[input_image],
        outputs=[emotion_scores, analysis_text, report_text]
    )
    
    download_btn.click(
        fn=lambda x: x,
        inputs=[report_text],
        outputs=[gr.File(label="Download Report")]
    )

app.launch()