Update app.py

app.py

@@ -1,88 +1,98 @@
 #!/usr/bin/env python3
 """
-Pure Flask App for Aphasia Classification
-No Gradio dependency - works anywhere!
+Lightweight Aphasia Classification App
+Optimized for Hugging Face Spaces with lazy loading and fallbacks
 """
 
-from flask import Flask, request, render_template_string, jsonify, send_from_directory
+from flask import Flask, request, render_template_string, jsonify
 import os
 import tempfile
 import logging
 import json
+import threading
+import time
 from pathlib import Path
-import traceback
 
 # Set up logging
-logging.basicConfig(level=logging.INFO)
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 logger = logging.getLogger(__name__)
 
 app = Flask(__name__)
-app.config['MAX_CONTENT_LENGTH'] = 100 * 1024 * 1024  # 100MB max
+app.config['MAX_CONTENT_LENGTH'] = 50 * 1024 * 1024  # 50MB max (reduced)
 
-print("🚀 Starting Aphasia Classification System (Flask)")
+print("🚀 Starting Lightweight Aphasia Classification System")
 
-def safe_import_modules():
-    """Import pipeline modules safely"""
-    modules = {}
+# Global state
+MODULES = {}
+MODELS_LOADED = False
+LOADING_STATUS = "Starting up..."
+
+def lazy_import_modules():
+    """Import modules only when needed"""
+    global MODULES, MODELS_LOADED, LOADING_STATUS
     
-    try:
-        from utils_audio import convert_to_wav
-        modules['convert_to_wav'] = convert_to_wav
-        logger.info("✓ utils_audio imported")
-    except Exception as e:
-        logger.error(f"✗ utils_audio failed: {e}")
-        modules['convert_to_wav'] = None
+    if MODELS_LOADED:
+        return True
     
     try:
+        LOADING_STATUS = "Loading audio processing..."
+        logger.info("Importing utils_audio...")
+        from utils_audio import convert_to_wav
+        MODULES['convert_to_wav'] = convert_to_wav
+        logger.info("✓ Audio processing loaded")
+        
+        LOADING_STATUS = "Loading speech analysis..."
+        logger.info("Importing to_cha...")
         from to_cha import to_cha_from_wav
-        modules['to_cha_from_wav'] = to_cha_from_wav
-        logger.info("✓ to_cha imported")
-    except Exception as e:
-        logger.error(f"✗ to_cha failed: {e}")
-        modules['to_cha_from_wav'] = None
-    
-    try:
+        MODULES['to_cha_from_wav'] = to_cha_from_wav
+        logger.info("✓ Speech analysis loaded")
+        
+        LOADING_STATUS = "Loading data conversion..."
+        logger.info("Importing cha_json...")
         from cha_json import cha_to_json_file
-        modules['cha_to_json_file'] = cha_to_json_file
-        logger.info("✓ cha_json imported")
-    except Exception as e:
-        logger.error(f"✗ cha_json failed: {e}")
-        modules['cha_to_json_file'] = None
-    
-    try:
+        MODULES['cha_to_json_file'] = cha_to_json_file
+        logger.info("✓ Data conversion loaded")
+        
+        LOADING_STATUS = "Loading AI model..."
+        logger.info("Importing output...")
         from output import predict_from_chajson
-        modules['predict_from_chajson'] = predict_from_chajson
-        logger.info("✓ output imported")
+        MODULES['predict_from_chajson'] = predict_from_chajson
+        logger.info("✓ AI model loaded")
+        
+        MODELS_LOADED = True
+        LOADING_STATUS = "Ready!"
+        logger.info("🎉 All modules loaded successfully!")
+        return True
+        
     except Exception as e:
-        logger.error(f"✗ output failed: {e}")
-        modules['predict_from_chajson'] = None
-    
-    return modules
+        logger.error(f"Failed to load modules: {e}")
+        LOADING_STATUS = f"Error: {str(e)}"
+        return False
+
+def background_loader():
+    """Load modules in background thread"""
+    logger.info("Starting background module loading...")
+    lazy_import_modules()
 
-# Import modules
-MODULES = safe_import_modules()
-MODEL_DIR = "."
+# Start loading modules in background
+loading_thread = threading.Thread(target=background_loader, daemon=True)
+loading_thread.start()
 
-# HTML Template
+# HTML Template (simplified)
 HTML_TEMPLATE = """
 <!DOCTYPE html>
 <html lang="en">
 <head>
     <meta charset="UTF-8">
     <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>🧠 Aphasia Classification System</title>
+    <title>🧠 Aphasia Classification</title>
     <style>
-        * {
-            margin: 0;
-            padding: 0;
-            box-sizing: border-box;
-        }
-        
         body {
             font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif;
             background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
             min-height: 100vh;
             padding: 20px;
+            margin: 0;
         }
         
         .container {
@@ -101,17 +111,6 @@ HTML_TEMPLATE = """
             text-align: center;
         }
         
-        .header h1 {
-            font-size: 2.5em;
-            margin-bottom: 10px;
-            font-weight: 700;
-        }
-        
-        .header p {
-            font-size: 1.1em;
-            opacity: 0.9;
-        }
-        
         .content {
             padding: 40px 30px;
         }
@@ -124,24 +123,20 @@ HTML_TEMPLATE = """
             border-left: 4px solid #28a745;
         }
         
-        .status h3 {
-            color: #28a745;
-            margin-bottom: 10px;
+        .status.loading {
+            border-left-color: #ffc107;
+        }
+        
+        .status.error {
+            border-left-color: #dc3545;
         }
         
         .upload-section {
             background: #f8f9fa;
             border-radius: 15px;
             padding: 30px;
-            margin-bottom: 30px;
-            border: 2px dashed #dee2e6;
             text-align: center;
-            transition: all 0.3s ease;
-        }
-        
-        .upload-section:hover {
-            border-color: #667eea;
-            background: #f0f4ff;
+            margin-bottom: 30px;
         }
         
         .file-input {
@@ -155,7 +150,6 @@ HTML_TEMPLATE = """
             padding: 15px 30px;
             border-radius: 50px;
             cursor: pointer;
-            font-size: 1.1em;
             font-weight: 600;
             transition: transform 0.2s ease;
         }
@@ -170,22 +164,15 @@ HTML_TEMPLATE = """
             border: none;
             padding: 15px 40px;
             border-radius: 50px;
-            font-size: 1.1em;
             font-weight: 600;
             cursor: pointer;
             margin-top: 20px;
             transition: all 0.2s ease;
         }
         
-        .analyze-btn:hover {
-            background: #218838;
-            transform: translateY(-2px);
-        }
-        
         .analyze-btn:disabled {
             background: #6c757d;
             cursor: not-allowed;
-            transform: none;
         }
         
         .results {
@@ -194,15 +181,8 @@ HTML_TEMPLATE = """
             padding: 30px;
             margin-top: 30px;
             display: none;
-        }
-        
-        .results.success {
-            border-left: 4px solid #28a745;
-        }
-        
-        .results.error {
-            border-left: 4px solid #dc3545;
-            background: #fff5f5;
+            white-space: pre-wrap;
+            font-family: monospace;
         }
         
         .loading {
@@ -226,30 +206,14 @@ HTML_TEMPLATE = """
             100% { transform: rotate(360deg); }
         }
         
-        .supported-formats {
-            text-align: center;
-            color: #6c757d;
-            margin-top: 15px;
-            font-size: 0.9em;
-        }
-        
-        .about {
-            background: #fff;
-            border-radius: 15px;
-            padding: 30px;
-            margin-top: 30px;
-            border: 1px solid #dee2e6;
-        }
-        
-        .about h3 {
-            color: #333;
-            margin-bottom: 15px;
-        }
-        
-        .about p {
-            color: #666;
-            line-height: 1.6;
-            margin-bottom: 10px;
+        .refresh-btn {
+            background: #17a2b8;
+            color: white;
+            border: none;
+            padding: 10px 20px;
+            border-radius: 25px;
+            cursor: pointer;
+            margin-left: 10px;
         }
     </style>
 </head>
@@ -257,18 +221,19 @@ HTML_TEMPLATE = """
     <div class="container">
         <div class="header">
             <h1>🧠 Aphasia Classification</h1>
-            <p>Advanced AI-powered speech analysis for aphasia type identification</p>
+            <p>AI-powered speech analysis for aphasia identification</p>
         </div>
         
         <div class="content">
-            <div class="status">
-                <h3>{{ status_title }}</h3>
-                <div>{{ status_details | safe }}</div>
+            <div class="status" id="statusBox">
+                <h3 id="statusTitle">🔄 System Status</h3>
+                <div id="statusText">{{ status_message }}</div>
+                <button class="refresh-btn" onclick="checkStatus()">Refresh Status</button>
             </div>
             
             <div class="upload-section">
                 <h3>📁 Upload Audio File</h3>
-                <p>Select an audio file containing speech for analysis</p>
+                <p>Upload speech audio for aphasia classification</p>
                 
                 <form id="uploadForm" enctype="multipart/form-data">
                     <input type="file" id="audioFile" name="audio" class="file-input" accept="audio/*" required>
@@ -281,47 +246,72 @@ HTML_TEMPLATE = """
                     </button>
                 </form>
                 
-                <div class="supported-formats">
-                    Supported: MP3, WAV, MP4, M4A, FLAC, OGG
-                </div>
+                <p style="color: #666; margin-top: 15px; font-size: 0.9em;">
+                    Supported: MP3, WAV, M4A (max 50MB)
+                </p>
             </div>
             
             <div class="loading" id="loading">
                 <div class="spinner"></div>
-                <h3>🔄 Analyzing Audio...</h3>
-                <p>This may take 1-3 minutes depending on file size</p>
+                <h3>🔄 Processing Audio...</h3>
+                <p>This may take 2-5 minutes. Please be patient.</p>
             </div>
             
-            <div class="results" id="results">
-                <div id="resultsContent"></div>
-            </div>
-            
-            <div class="about">
-                <h3>About This System</h3>
-                <p>This AI system analyzes speech patterns to classify different types of aphasia, including:</p>
-                <p><strong>• Broca's Aphasia:</strong> Non-fluent speech with preserved comprehension</p>
-                <p><strong>• Wernicke's Aphasia:</strong> Fluent but often meaningless speech</p>
-                <p><strong>• Anomic Aphasia:</strong> Word-finding difficulties</p>
-                <p><strong>• Conduction Aphasia:</strong> Fluent speech with poor repetition</p>
-                <p><strong>• Global Aphasia:</strong> Severe impairment in all language areas</p>
-                <br>
-                <p><em>Note: This tool is for research and educational purposes. Always consult healthcare professionals for clinical decisions.</em></p>
-            </div>
+            <div class="results" id="results"></div>
         </div>
     </div>
 
     <script>
+        // Check status periodically
+        function checkStatus() {
+            fetch('/status')
+                .then(response => response.json())
+                .then(data => {
+                    const statusBox = document.getElementById('statusBox');
+                    const statusTitle = document.getElementById('statusTitle');
+                    const statusText = document.getElementById('statusText');
+                    
+                    if (data.ready) {
+                        statusBox.className = 'status';
+                        statusTitle.textContent = '🟢 System Ready';
+                        statusText.textContent = 'All components loaded. Ready to process audio files.';
+                    } else {
+                        statusBox.className = 'status loading';
+                        statusTitle.textContent = '🟡 Loading...';
+                        statusText.textContent = data.status;
+                    }
+                })
+                .catch(error => {
+                    const statusBox = document.getElementById('statusBox');
+                    statusBox.className = 'status error';
+                    document.getElementById('statusTitle').textContent = '🔴 Error';
+                    document.getElementById('statusText').textContent = 'Failed to check status';
+                });
+        }
+        
+        // Check status every 5 seconds
+        setInterval(checkStatus, 5000);
+        
+        // Form submission
         document.getElementById('uploadForm').addEventListener('submit', async function(e) {
             e.preventDefault();
             
             const fileInput = document.getElementById('audioFile');
-            const analyzeBtn = document.getElementById('analyzeBtn');
             const loading = document.getElementById('loading');
             const results = document.getElementById('results');
-            const resultsContent = document.getElementById('resultsContent');
+            const analyzeBtn = document.getElementById('analyzeBtn');
             
             if (!fileInput.files[0]) {
-                alert('Please select an audio file first');
+                alert('Please select an audio file');
+                return;
+            }
+            
+            // Check if system is ready
+            const statusCheck = await fetch('/status');
+            const status = await statusCheck.json();
+            
+            if (!status.ready) {
+                alert('System is still loading. Please wait and try again.');
                 return;
             }
             
@@ -329,7 +319,7 @@ HTML_TEMPLATE = """
             loading.style.display = 'block';
             results.style.display = 'none';
             analyzeBtn.disabled = true;
-            analyzeBtn.textContent = '🔄 Processing...';
+            analyzeBtn.textContent = 'Processing...';
             
             try {
                 const formData = new FormData();
@@ -342,32 +332,30 @@ HTML_TEMPLATE = """
                 
                 const data = await response.json();
                 
-                // Hide loading
                 loading.style.display = 'none';
                 
                 if (data.success) {
-                    resultsContent.innerHTML = '<pre style="white-space: pre-wrap; font-family: inherit;">' + data.result + '</pre>';
-                    results.className = 'results success';
+                    results.textContent = data.result;
+                    results.style.borderLeft = '4px solid #28a745';
                 } else {
-                    resultsContent.innerHTML = '<h3 style="color: #dc3545;">❌ Error</h3><p>' + data.error + '</p>';
-                    results.className = 'results error';
+                    results.textContent = 'Error: ' + data.error;
+                    results.style.borderLeft = '4px solid #dc3545';
                 }
                 
                 results.style.display = 'block';
                 
             } catch (error) {
                 loading.style.display = 'none';
-                resultsContent.innerHTML = '<h3 style="color: #dc3545;">❌ Network Error</h3><p>Failed to process request: ' + error.message + '</p>';
-                results.className = 'results error';
+                results.textContent = 'Network error: ' + error.message;
+                results.style.borderLeft = '4px solid #dc3545';
                 results.style.display = 'block';
             }
             
-            // Reset button
             analyzeBtn.disabled = false;
             analyzeBtn.textContent = '🔍 Analyze Speech';
         });
         
-        // Update file label when file is selected
+        // File selection feedback
         document.getElementById('audioFile').addEventListener('change', function(e) {
             const label = document.querySelector('.file-label');
             if (e.target.files[0]) {
@@ -384,26 +372,29 @@ HTML_TEMPLATE = """
 @app.route('/')
 def index():
     """Main page"""
-    # Check system status
-    modules_ready = all(MODULES.values())
-    missing_modules = [k for k, v in MODULES.items() if v is None]
-    
-    if modules_ready:
-        status_title = "🟢 System Ready"
-        status_details = "All components loaded successfully. Ready to process audio files."
-    else:
-        status_title = "🟡 System Loading"
-        status_details = f"Missing modules: {', '.join(missing_modules)}<br>Some features may not be available."
-    
-    return render_template_string(HTML_TEMPLATE, 
-                                status_title=status_title,
-                                status_details=status_details)
+    return render_template_string(HTML_TEMPLATE, status_message=LOADING_STATUS)
+
+@app.route('/status')
+def status():
+    """Status check endpoint"""
+    return jsonify({
+        'ready': MODELS_LOADED,
+        'status': LOADING_STATUS,
+        'modules_loaded': len(MODULES)
+    })
 
 @app.route('/analyze', methods=['POST'])
 def analyze_audio():
-    """Process uploaded audio file"""
+    """Process uploaded audio - only if models are loaded"""
     try:
-        # Check if file was uploaded
+        # Check if system is ready
+        if not MODELS_LOADED:
+            return jsonify({
+                'success': False, 
+                'error': f'System still loading: {LOADING_STATUS}'
+            })
+        
+        # Check file upload
         if 'audio' not in request.files:
             return jsonify({'success': False, 'error': 'No audio file uploaded'})
         
@@ -411,37 +402,32 @@ def analyze_audio():
         if audio_file.filename == '':
             return jsonify({'success': False, 'error': 'No file selected'})
         
-        # Check if modules are available
-        if not all(MODULES.values()):
-            missing = [k for k, v in MODULES.items() if v is None]
-            return jsonify({'success': False, 'error': f'System not ready. Missing: {", ".join(missing)}'})
-        
-        # Save uploaded file temporarily
+        # Save uploaded file
         with tempfile.NamedTemporaryFile(delete=False, suffix=os.path.splitext(audio_file.filename)[1]) as tmp_file:
             audio_file.save(tmp_file.name)
-            temp_audio_path = tmp_file.name
+            temp_path = tmp_file.name
         
         try:
-            logger.info("🎵 Starting audio processing pipeline...")
+            logger.info("🎵 Starting audio processing...")
             
             # Step 1: Convert to WAV
-            wav_path = MODULES['convert_to_wav'](temp_audio_path, sr=16000, mono=True)
-            logger.info("✓ Audio converted to WAV")
+            logger.info("Converting to WAV...")
+            wav_path = MODULES['convert_to_wav'](temp_path, sr=16000, mono=True)
             
-            # Step 2: Generate CHA file
+            # Step 2: Generate CHA
+            logger.info("Generating CHA file...")
             cha_path = MODULES['to_cha_from_wav'](wav_path, lang="eng")
-            logger.info("✓ CHA file generated")
             
-            # Step 3: Convert CHA to JSON
+            # Step 3: Convert to JSON
+            logger.info("Converting to JSON...")
             json_path, _ = MODULES['cha_to_json_file'](cha_path)
-            logger.info("✓ JSON conversion completed")
             
-            # Step 4: Run classification
-            results = MODULES['predict_from_chajson'](MODEL_DIR, json_path, output_file=None)
-            logger.info("✓ Classification completed")
+            # Step 4: Classification
+            logger.info("Running classification...")
+            results = MODULES['predict_from_chajson'](".", json_path, output_file=None)
             
-            # Clean up temporary files
-            for temp_file in [temp_audio_path, wav_path, cha_path, json_path]:
+            # Cleanup
+            for temp_file in [temp_path, wav_path, cha_path, json_path]:
                 try:
                     os.unlink(temp_file)
                 except:
@@ -451,79 +437,53 @@ def analyze_audio():
             if "predictions" in results and results["predictions"]:
                 pred = results["predictions"][0]
                 
-                if "error" in pred:
-                    return jsonify({'success': False, 'error': f'Classification error: {pred["error"]}'})
-                
-                # Format the result text
                 classification = pred["prediction"]["predicted_class"]
                 confidence = pred["prediction"]["confidence_percentage"]
-                type_name = pred["class_description"]["name"]
-                description = pred["class_description"]["description"]
+                description = pred["class_description"]["name"]
                 severity = pred["additional_predictions"]["predicted_severity_level"]
                 fluency = pred["additional_predictions"]["fluency_rating"]
                 
                 result_text = f"""🧠 APHASIA CLASSIFICATION RESULTS
 
-🎯 Primary Classification: {classification}
+🎯 Classification: {classification}
 📊 Confidence: {confidence}
-📋 Type: {type_name}
-📈 Severity Level: {severity}/3
-🗣️ Fluency Rating: {fluency}
+📋 Type: {description}
+📈 Severity: {severity}/3
+🗣️ Fluency: {fluency}
 
-📊 Top 3 Probability Rankings:"""
+📊 Top 3 Probabilities:"""
 
-                # Add probability distribution
                 prob_dist = pred["probability_distribution"]
                 for i, (atype, info) in enumerate(list(prob_dist.items())[:3], 1):
                     result_text += f"\n{i}. {atype}: {info['percentage']}"
                 
                 result_text += f"""
 
-📝 Clinical Description:
-{description}
+📝 Description:
+{pred["class_description"]["description"]}
 
-📊 Processing Summary:
-• Total sentences analyzed: {results.get('total_sentences', 'N/A')}
-• Average confidence: {results.get('summary', {}).get('average_confidence', 'N/A')}
-• Processing completed successfully
+✅ Processing completed successfully!
 """
                 
                 return jsonify({'success': True, 'result': result_text})
             else:
-                return jsonify({'success': False, 'error': 'No predictions generated from the audio file'})
+                return jsonify({'success': False, 'error': 'No predictions generated'})
                 
         except Exception as e:
-            # Clean up temp file on error
+            # Cleanup on error
             try:
-                os.unlink(temp_audio_path)
+                os.unlink(temp_path)
             except:
                 pass
             raise e
             
     except Exception as e:
         logger.error(f"Processing error: {e}")
-        traceback.print_exc()
-        return jsonify({'success': False, 'error': f'Processing failed: {str(e)}'})
-
-@app.route('/health')
-def health_check():
-    """Health check endpoint"""
-    modules_ready = all(MODULES.values())
-    return jsonify({
-        'status': 'healthy' if modules_ready else 'degraded',
-        'modules': {k: v is not None for k, v in MODULES.items()},
-        'ready': modules_ready
-    })
+        return jsonify({'success': False, 'error': str(e)})
 
 if __name__ == '__main__':
-    # Get port from environment (for Hugging Face Spaces)
     port = int(os.environ.get('PORT', 7860))
-    host = os.environ.get('HOST', '0.0.0.0')
-    
-    print(f"🚀 Starting Flask app on {host}:{port}")
-    print("📋 Modules status:")
-    for name, module in MODULES.items():
-        status = "✓" if module else "❌"
-        print(f"   {status} {name}")
+    print(f"🚀 Starting on port {port}")
+    print("🔄 Models loading in background...")
     
-    app.run(host=host, port=port, debug=False)
+    app.run(host='0.0.0.0', port=port, debug=False, threaded=True)