Update app.py

app.py

@@ -6,92 +6,36 @@ from huggingface_hub import hf_hub_download
 import time
 import os
 
-# Configuración de GPU para TensorFlow
-physical_devices = tf.config.list_physical_devices('GPU')
-if physical_devices:
-    print("GPU disponible. Configurando...")
-    try:
-        for gpu in physical_devices:
-            tf.config.experimental.set_memory_growth(gpu, True)
-        print("Configuración de GPU completada")
-    except Exception as e:
-        print(f"Error en configuración de GPU: {e}")
-else:
-    print("No se detectó GPU. El procesamiento será más lento.")
+model_path = hf_hub_download(repo_id="Bmo411/DenoisingAutoencoder", filename="autoencoder_complete_model_Fourier.keras")
+model = tf.keras.models.load_model(model_path)
+
+from tensorflow.keras.saving import register_keras_serializable
 
-# Definir funciones personalizadas para las transformadas de Fourier
+@register_keras_serializable()
 def fourier_transform(x):
     fourier = tf.signal.fft2d(tf.cast(x, tf.complex64))
     fourier = tf.complex(tf.math.real(fourier), tf.math.imag(fourier))
     fourier = tf.abs(fourier)
     return tf.concat([tf.math.real(fourier), tf.math.imag(fourier)], axis=-1)
 
+@register_keras_serializable()
 def inverse_fourier_transform(x):
     real_part, imag_part = tf.split(x, num_or_size_splits=2, axis=-1)
     complex_fourier = tf.complex(real_part, imag_part)
     return tf.abs(tf.signal.ifft2d(complex_fourier))
-
-# Construir modelo manualmente
-def build_autoencoder(input_shape=(256, 256, 3)):
-    """Reconstruir el modelo autoencoder manualmente"""
-    
-    # Definir entradas
-    inputs = tf.keras.layers.Input(shape=input_shape)
-    
-    # Aplicar transformada de Fourier (opcional)
-    # x = tf.keras.layers.Lambda(fourier_transform)(inputs)
-    x = inputs  # Skip Fourier transform for now
-    
-    # Encoder
-    x = tf.keras.layers.Conv2D(32, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.MaxPooling2D((2, 2), padding='same')(x)
-    x = tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.MaxPooling2D((2, 2), padding='same')(x)
-    x = tf.keras.layers.Conv2D(128, (3, 3), activation='relu', padding='same')(x)
-    encoded = tf.keras.layers.MaxPooling2D((2, 2), padding='same')(x)
-    
-    # Decoder
-    x = tf.keras.layers.Conv2D(128, (3, 3), activation='relu', padding='same')(encoded)
-    x = tf.keras.layers.UpSampling2D((2, 2))(x)
-    x = tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.UpSampling2D((2, 2))(x)
-    x = tf.keras.layers.Conv2D(32, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.UpSampling2D((2, 2))(x)
-    
-    # Output
-    # x = tf.keras.layers.Lambda(inverse_fourier_transform)(x)  # Skip inverse transform
-    outputs = tf.keras.layers.Conv2D(3, (3, 3), activation='sigmoid', padding='same')(x)
-    
-    # Crear modelo
-    model = tf.keras.models.Model(inputs, outputs)
     
-    return model
-
-# Opción alternativa: Crear un modelo simplificado
-def build_simple_autoencoder(input_shape=(256, 256, 3)):
-    """Crear un autoencoder simple (sin Fourier)"""
-    
-    inputs = tf.keras.layers.Input(shape=input_shape)
-    
-    # Encoder
-    x = tf.keras.layers.Conv2D(32, (3, 3), activation='relu', padding='same')(inputs)
-    x = tf.keras.layers.MaxPooling2D((2, 2), padding='same')(x)
-    x = tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.MaxPooling2D((2, 2), padding='same')(x)
-    
-    # Decoder
-    x = tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.UpSampling2D((2, 2))(x)
-    x = tf.keras.layers.Conv2D(32, (3, 3), activation='relu', padding='same')(x)
-    x = tf.keras.layers.UpSampling2D((2, 2))(x)
-    
-    # Output
-    outputs = tf.keras.layers.Conv2D(3, (3, 3), activation='sigmoid', padding='same')(x)
-    
-    # Crear modelo
-    model = tf.keras.models.Model(inputs, outputs)
-    
-    return model
+# Configuración de GPU para TensorFlow
+physical_devices = tf.config.list_physical_devices('GPU')
+if physical_devices:
+    print("GPU disponible. Configurando...")
+    try:
+        for gpu in physical_devices:
+            tf.config.experimental.set_memory_growth(gpu, True)
+        print("Configuración de GPU completada")
+    except Exception as e:
+        print(f"Error en configuración de GPU: {e}")
+else:
+    print("No se detectó GPU. El procesamiento será más lento.")
 
 # Funciones de preprocesamiento optimizadas
 def degrade_image(image, downscale_factor=4):
@@ -138,10 +82,6 @@ def preprocess_image(image, std_dev=0.1, downscale_factor=4, target_size=(256, 2
 
     return noisy_img
 
-print("Crear modelo simplificado...")
-model = build_simple_autoencoder()
-print("Modelo creado correctamente")
-
 # Función de denoising
 def Denoiser(imagen):
     """Aplica el modelo autoencoder para eliminar ruido de la imagen."""