Update app.py

app.py

@@ -333,9 +333,11 @@ def create_probability_chart(predictions, consensus_class):
         if predictions:
             # Obtener probabilidades promedio
             avg_probs = np.zeros(7)
-            for pred in predictions:
+            valid_predictions = [p for p in predictions if p.get('success', False)]
+            
+            for pred in valid_predictions:
                 avg_probs += pred['probabilities']
-            avg_probs /= len(predictions)
+            avg_probs /= len(valid_predictions)
             
             colors = ['#ff6b35' if i in MALIGNANT_INDICES else '#44ff44' for i in range(7)]
             bars = ax1.bar(range(7), avg_probs, color=colors, alpha=0.8)
@@ -360,16 +362,17 @@ def create_probability_chart(predictions, consensus_class):
                         f'{height:.2%}', ha='center', va='bottom', fontsize=9)
         
         # Gráfico 2: Confianza por modelo
-        model_names = [pred['model'].split(' ')[1] if len(pred['model'].split(' ')) > 1 else pred['model'] for pred in predictions]
-        confidences = [pred['confidence'] for pred in predictions]
+        valid_predictions = [p for p in predictions if p.get('success', False)]
+        model_names = [pred['model'].split(' ')[1] if len(pred['model'].split(' ')) > 1 else pred['model'] for pred in valid_predictions]
+        confidences = [pred['confidence'] for pred in valid_predictions]
         
-        colors_conf = ['#ff6b35' if pred['is_malignant'] else '#44ff44' for pred in predictions]
-        bars2 = ax2.bar(range(len(predictions)), confidences, color=colors_conf, alpha=0.8)
+        colors_conf = ['#ff6b35' if pred['is_malignant'] else '#44ff44' for pred in valid_predictions]
+        bars2 = ax2.bar(range(len(valid_predictions)), confidences, color=colors_conf, alpha=0.8)
         
         ax2.set_xlabel('Modelos')
         ax2.set_ylabel('Confianza')
         ax2.set_title('🎯 Confianza por Modelo')
-        ax2.set_xticks(range(len(predictions)))
+        ax2.set_xticks(range(len(valid_predictions)))
         ax2.set_xticklabels(model_names, rotation=45)
         ax2.grid(True, alpha=0.3)
         ax2.set_ylim(0, 1)
@@ -394,8 +397,70 @@ def create_probability_chart(predictions, consensus_class):
     except Exception as e:
         print(f"Error creando gráfico: {e}")
         return "<p>❌ Error generando gráfico de probabilidades</p>"
-    
+
+def create_heatmap(predictions):
+    """Crear mapa de calor de probabilidades por modelo"""
     try:
+        valid_predictions = [p for p in predictions if p.get('success', False)]
+        
+        if not valid_predictions:
+            return "<p>No hay datos suficientes para el mapa de calor</p>"
+        
+        # Crear matriz de probabilidades
+        prob_matrix = np.array([pred['probabilities'] for pred in valid_predictions])
+        
+        # Crear figura
+        fig, ax = plt.subplots(figsize=(10, 6))
+        
+        # Crear mapa de calor
+        im = ax.imshow(prob_matrix, cmap='RdYlGn_r', aspect='auto', vmin=0, vmax=1)
+        
+        # Configurar etiquetas
+        ax.set_xticks(np.arange(7))
+        ax.set_yticks(np.arange(len(valid_predictions)))
+        ax.set_xticklabels([cls.split('(')[1].rstrip(')') for cls in CLASSES])
+        ax.set_yticklabels([pred['model'] for pred in valid_predictions])
+        
+        # Rotar etiquetas del eje x
+        plt.setp(ax.get_xticklabels(), rotation=45, ha="right", rotation_mode="anchor")
+        
+        # Añadir valores en las celdas
+        for i in range(len(valid_predictions)):
+            for j in range(7):
+                text = ax.text(j, i, f'{prob_matrix[i, j]:.2f}',
+                             ha="center", va="center", color="white" if prob_matrix[i, j] > 0.5 else "black",
+                             fontsize=8)
+        
+        ax.set_title("Mapa de Calor: Probabilidades por Modelo y Clase")
+        fig.tight_layout()
+        
+        # Añadir barra de color
+        cbar = plt.colorbar(im, ax=ax)
+        cbar.set_label('Probabilidad', rotation=270, labelpad=15)
+        
+        # Convertir a base64
+        buf = io.BytesIO()
+        plt.savefig(buf, format='png', dpi=300, bbox_inches='tight')
+        buf.seek(0)
+        heatmap_b64 = base64.b64encode(buf.getvalue()).decode()
+        plt.close()
+        
+        return f'<img src="data:image/png;base64,{heatmap_b64}" style="width:100%; max-width:800px;">'
+        
+    except Exception as e:
+        print(f"Error creando mapa de calor: {e}")
+        return "<p>❌ Error generando mapa de calor</p>"
+
+def analizar_lesion(img):
+    """Función principal para analizar la lesión"""
+    try:
+        if img is None:
+            return "<h3>⚠️ Por favor, carga una imagen</h3>"
+        
+        # Verificar que hay modelos cargados
+        if not loaded_models or all(m.get('type') == 'dummy' for m in loaded_models.values()):
+            return "<h3>❌ Error del Sistema</h3><p>No hay modelos disponibles. Por favor, recarga la aplicación.</p>"
+        
         # Convertir a RGB si es necesario
         if img.mode != 'RGB':
             img = img.convert('RGB')
@@ -404,9 +469,10 @@ def create_probability_chart(predictions, consensus_class):
         
         # Obtener predicciones de todos los modelos cargados
         for model_name, model_data in loaded_models.items():
-            pred = predict_with_model(img, model_data)
-            if pred.get('success', False):
-                predictions.append(pred)
+            if model_data.get('type') != 'dummy':
+                pred = predict_with_model(img, model_data)
+                if pred.get('success', False):
+                    predictions.append(pred)
         
         if not predictions:
             return "<h3>❌ Error</h3><p>No se pudieron obtener predicciones de ningún modelo.</p>"
@@ -547,125 +613,10 @@ def create_probability_chart(predictions, consensus_class):
         </div>
         """
         
-        # Generar visualizaciones
-        probability_chart = create_probability_chart(predictions, consensus_class)
-        heatmap = create_heatmap(predictions)
-        
-        # Generar HTML del reporte COMPLETO
-        html_report = f"""
-        <div style="font-family: Arial, sans-serif; max-width: 1200px; margin: 0 auto;">
-            <h2 style="color: #2c3e50; text-align: center;">🏥 Análisis Completo de Lesión Cutánea</h2>
-            
-            <div style="background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); color: white; padding: 20px; border-radius: 10px; margin: 20px 0;">
-                <h3 style="margin: 0; text-align: center;">📋 Resultado de Consenso</h3>
-                <p style="font-size: 18px; text-align: center; margin: 10px 0;"><strong>{consensus_class}</strong></p>
-                <p style="text-align: center; margin: 5px 0;">Confianza Promedio: <strong>{avg_confidence:.1%}</strong></p>
-                <p style="text-align: center; margin: 5px 0;">Consenso: <strong>{class_votes[consensus_class]}/{len(predictions)} modelos</strong></p>
-            </div>
-            
-            <div style="background: {risk_info['color']}; color: white; padding: 15px; border-radius: 8px; margin: 15px 0;">
-                <h4 style="margin: 0;">⚠️ Nivel de Riesgo: {risk_info['level']}</h4>
-                <p style="margin: 5px 0;"><strong>{risk_info['urgency']}</strong></p>
-                <p style="margin: 5px 0;">Tipo: {'🔴 Potencialmente maligna' if is_malignant else '🟢 Probablemente benigna'}</p>
-            </div>
-            
-            <div style="background: #e3f2fd; padding: 15px; border-radius: 8px; margin: 15px 0;">
-                <h4 style="color: #1976d2;">🤖 Resultados Individuales por Modelo</h4>
-        """
-        
-        # RESULTADOS INDIVIDUALES DETALLADOS
-        for i, pred in enumerate(predictions, 1):
-            if pred['success']:
-                model_risk = RISK_LEVELS[pred['predicted_idx']]
-                malignant_status = "🔴 Maligna" if pred['is_malignant'] else "🟢 Benigna"
-                
-                html_report += f"""
-                <div style="margin: 15px 0; padding: 15px; background: white; border-radius: 8px; border-left: 5px solid {'#ff6b35' if pred['is_malignant'] else '#44ff44'}; box-shadow: 0 2px 4px rgba(0,0,0,0.1);">
-                    <div style="display: flex; justify-content: space-between; align-items: center; margin-bottom: 10px;">
-                        <h5 style="margin: 0; color: #333;">#{i}. {pred['model']}</h5>
-                        <span style="background: {model_risk['color']}; color: white; padding: 4px 8px; border-radius: 4px; font-size: 12px;">{model_risk['level']}</span>
-                    </div>
-                    
-                    <div style="display: grid; grid-template-columns: 1fr 1fr 1fr; gap: 10px; font-size: 14px;">
-                        <div><strong>Diagnóstico:</strong><br>{pred['class']}</div>
-                        <div><strong>Confianza:</strong><br>{pred['confidence']:.1%}</div>
-                        <div><strong>Clasificación:</strong><br>{malignant_status}</div>
-                    </div>
-                    
-                    <div style="margin-top: 10px;">
-                        <strong>Top 3 Probabilidades:</strong><br>
-                        <div style="font-size: 12px; color: #666;">
-                """
-                
-                # Top 3 probabilidades para este modelo
-                top_indices = np.argsort(pred['probabilities'])[-3:][::-1]
-                for idx in top_indices:
-                    prob = pred['probabilities'][idx]
-                    if prob > 0.01:  # Solo mostrar si > 1%
-                        html_report += f"• {CLASSES[idx].split('(')[1].rstrip(')')}: {prob:.1%}<br>"
-                
-                html_report += f"""
-                        </div>
-                        <div style="margin-top: 8px; font-size: 12px; color: #888;">
-                            <strong>Recomendación:</strong> {model_risk['urgency']}
-                        </div>
-                    </div>
-                </div>
-                """
-            else:
-                html_report += f"""
-                <div style="margin: 10px 0; padding: 10px; background: #ffebee; border-radius: 5px; border-left: 4px solid #f44336;">
-                    <strong>❌ {pred['model']}</strong><br>
-                    <span style="color: #d32f2f;">Error: {pred.get('error', 'Desconocido')}</span>
-                </div>
-                """
-        
-        html_report += f"""
-            </div>
-            
-            <div style="background: #f8f9fa; padding: 15px; border-radius: 8px; margin: 15px 0;">
-                <h4 style="color: #495057;">📊 Análisis Estadístico</h4>
-                <div style="display: grid; grid-template-columns: repeat(auto-fit, minmax(200px, 1fr)); gap: 15px;">
-                    <div>
-                        <strong>Modelos Activos:</strong> {len([p for p in predictions if p['success']])}/{len(predictions)}<br>
-                        <strong>Acuerdo Total:</strong> {class_votes[consensus_class]}/{len([p for p in predictions if p['success']])}<br>
-                        <strong>Confianza Máxima:</strong> {max([p['confidence'] for p in predictions if p['success']]):.1%}
-                    </div>
-                    <div>
-                        <strong>Diagnósticos Malignos:</strong> {len([p for p in predictions if p.get('success') and p.get('is_malignant')])}<br>
-                        <strong>Diagnósticos Benignos:</strong> {len([p for p in predictions if p.get('success') and not p.get('is_malignant')])}<br>
-                        <strong>Consenso Maligno:</strong> {'Sí' if is_malignant else 'No'}
-                    </div>
-                </div>
-            </div>
-            
-            <div style="background: #ffffff; padding: 15px; border-radius: 8px; margin: 15px 0; border: 1px solid #ddd;">
-                <h4 style="color: #333;">📈 Gráficos de Análisis</h4>
-                {probability_chart}
-            </div>
-            
-            <div style="background: #ffffff; padding: 15px; border-radius: 8px; margin: 15px 0; border: 1px solid #ddd;">
-                <h4 style="color: #333;">🔥 Mapa de Calor de Probabilidades</h4>
-                {heatmap}
-            </div>
-            
-            <div style="background: #fff3e0; padding: 15px; border-radius: 8px; margin: 15px 0; border: 1px solid #ff9800;">
-                <h4 style="color: #f57c00;">⚠️ Advertencia Médica</h4>
-                <p style="margin: 5px 0;">Este análisis es solo una herramienta de apoyo diagnóstico basada en IA.</p>
-                <p style="margin: 5px 0;"><strong>Siempre consulte con un dermatólogo profesional para un diagnóstico definitivo.</strong></p>
-                <p style="margin: 5px 0;">No utilice esta información como único criterio para decisiones médicas.</p>
-                <p style="margin: 5px 0;"><em>Los resultados individuales de cada modelo se muestran para transparencia y análisis comparativo.</em></p>
-            </div>
-        </div>
-        """
-        
         return html_report
         
     except Exception as e:
         return f"<h3>❌ Error en el análisis</h3><p>Error técnico: {str(e)}</p><p>Por favor, intente con otra imagen.</p>"
-        
-    except Exception as e:
-        return f"<h3>❌ Error en el análisis</h3><p>Error técnico: {str(e)}</p><p>Por favor, intente con otra imagen.</p>"
 
 # Configuración de Gradio
 def create_interface():