Update app.py

app.py

@@ -6,10 +6,10 @@ import base64
 import io
 from fastai.vision.all import *
 import tensorflow as tf
-from tensorflow import keras
 import zipfile
 import os
 import traceback
+from transformers import AutoImageProcessor, AutoModelForImageClassification
 
 # Descomprimir el modelo si no se ha descomprimido aún
 if not os.path.exists("saved_model"):
@@ -17,12 +17,10 @@ if not os.path.exists("saved_model"):
         zip_ref.extractall("saved_model")
 
 # Cargar modelo ISIC con TensorFlow desde el directorio correcto
-from keras.layers import TFSMLayer
-
 try:
-    model_isic = TFSMLayer("saved_model/saved_model", call_endpoint="serving_default")
+    model_isic = tf.saved_model.load("saved_model")
 except Exception as e:
-    print("\U0001F534 Error al cargar el modelo ISIC con TFSMLayer:", e)
+    print("\U0001F534 Error al cargar el modelo ISIC:", e)
     raise
 
 # Cargar modelos fastai
@@ -30,7 +28,6 @@ model_malignancy = load_learner("ada_learn_malben.pkl")
 model_norm2000 = load_learner("ada_learn_skin_norm2000.pkl")
 
 # Cargar modelo ViT
-from transformers import AutoImageProcessor, AutoModelForImageClassification
 feature_extractor = AutoImageProcessor.from_pretrained("nateraw/vit-skin-cancer")
 model_vit = AutoModelForImageClassification.from_pretrained("nateraw/vit-skin-cancer")
 
@@ -46,16 +43,15 @@ RISK_LEVELS = {
     6: {"label": "vasc", "color": "#073B4C", "weight": 0.4},
 }
 
-# Preprocesado para TensorFlow ISIC
 def preprocess_image_isic(pil_image):
     image = pil_image.resize((224, 224))
     array = np.array(image) / 255.0
     return np.expand_dims(array, axis=0)
 
-# Función de análisis
 def analizar_lesion_combined(img):
     try:
         img_fastai = PILImage.create(img)
+
         inputs = feature_extractor(img, return_tensors="pt")
         with torch.no_grad():
             outputs = model_vit(**inputs)
@@ -69,17 +65,17 @@ def analizar_lesion_combined(img):
         pred_fast_type, _, probs_fast_type = model_norm2000.predict(img_fastai)
 
         x_isic = preprocess_image_isic(img)
-        preds_isic_dict = model_isic(x_isic)
-        print("\U0001F50D Claves de salida de model_isic:", preds_isic_dict.keys())
-        key = list(preds_isic_dict.keys())[0]
-        preds_isic = preds_isic_dict[key].numpy()[0]
+        isic_func = model_isic.signatures["serving_default"]
+        preds_isic_tensor = isic_func(tf.constant(x_isic))
+        key = list(preds_isic_tensor.keys())[0]
+        preds_isic = preds_isic_tensor[key].numpy()[0]
         pred_idx_isic = int(np.argmax(preds_isic))
         pred_class_isic = CLASSES[pred_idx_isic]
         confidence_isic = preds_isic[pred_idx_isic]
 
         colors_bars = [RISK_LEVELS[i]['color'] for i in range(7)]
         fig, ax = plt.subplots(figsize=(8, 3))
-        ax.bar(CLASSES, probs_vit*100, color=colors_bars)
+        ax.bar(CLASSES, probs_vit * 100, color=colors_bars)
         ax.set_title("Probabilidad ViT por tipo de lesión")
         ax.set_ylabel("Probabilidad (%)")
         ax.set_xticks(np.arange(len(CLASSES)))
@@ -101,7 +97,7 @@ def analizar_lesion_combined(img):
             <tr><td>🧬 Fast.ai (clasificación)</td><td><b>{pred_fast_type}</b></td><td>N/A</td></tr>
             <tr><td>⚠️ Fast.ai (malignidad)</td><td><b>{"Maligno" if prob_malignant > 0.5 else "Benigno"}</b></td><td>{prob_malignant:.1%}</td></tr>
             <tr><td>🔬 ISIC TensorFlow</td><td><b>{pred_class_isic}</b></td><td>{confidence_isic:.1%}</td></tr>
-        </table><br><b>🧺 Recomendación automática:</b><br>"""
+        </table><br><b>🮺 Recomendación automática:</b><br>"
 
         cancer_risk_score = sum(probs_vit[i] * RISK_LEVELS[i]['weight'] for i in range(7))
         if prob_malignant > 0.7 or cancer_risk_score > 0.6:
@@ -122,7 +118,6 @@ def analizar_lesion_combined(img):
         traceback.print_exc()
         return f"<b>Error interno:</b> {str(e)}", ""
 
-# INTERFAZ
 demo = gr.Interface(
     fn=analizar_lesion_combined,
     inputs=gr.Image(type="pil", label="Sube una imagen de la lesión"),
@@ -132,7 +127,6 @@ demo = gr.Interface(
     flagging_mode="never"
 )
 
-# LANZAMIENTO
 if __name__ == "__main__":
     demo.launch()