metadata v4

app.py

@@ -1,5 +1,5 @@
 import numpy as np
-from PIL import Image
+from PIL import Image, UnidentifiedImageError
 import gradio as gr
 from deepface import DeepFace
 from datasets import load_dataset, DownloadConfig, Image as HfImage
@@ -7,6 +7,8 @@ import os
 import pickle
 from pathlib import Path
 import gc
+import requests
+from io import BytesIO
 
 # 🔑 Configurar token de Hugging Face
 HF_TOKEN = os.getenv("HF_TOKEN")
@@ -38,44 +40,57 @@ def build_database():
         print("📂 Cargando embeddings desde el archivo...")
         with open(EMBEDDINGS_FILE, 'rb') as f:
             return pickle.load(f)
-    
+
     print("🔄 Calculando embeddings (esto puede tomar unos minutos)...")
     database = []
     batch_size = 10
-    
+
     for i in range(0, len(dataset), batch_size):
         batch = dataset[i:i + batch_size]
         print(f"📦 Procesando lote {i // batch_size + 1}/{(len(dataset) + batch_size - 1) // batch_size}")
-        
+
         for j, item in enumerate(batch):
             try:
-                img = item["image"]  # Ya es un objeto PIL.Image
+                # Manejar ruta local o URL remota
+                img_data = item["image"]
+                if isinstance(img_data, str):
+                    response = requests.get(img_data)
+                    img = Image.open(BytesIO(response.content))
+                elif isinstance(img_data, dict) and "bytes" in img_data:
+                    img = Image.open(BytesIO(img_data["bytes"]))
+                elif isinstance(img_data, Image.Image):
+                    img = img_data
+                else:
+                    raise ValueError(f"Formato de imagen no soportado: {type(img_data)}")
+
                 img_processed = preprocess_image(img)
                 embedding = DeepFace.represent(
                     img_path=img_processed,
                     model_name="Facenet",
                     enforce_detection=False
                 )[0]["embedding"]
-                
+
                 database.append((f"image_{i+j}", img, embedding))
                 print(f"✅ Procesada imagen {i+j+1}/{len(dataset)}")
-                
-                # Liberar memoria
+
                 del img_processed
                 gc.collect()
-            
+
+            except UnidentifiedImageError:
+                print(f"❌ Imagen no válida en {i+j}: no se pudo identificar")
+                continue
             except Exception as e:
                 print(f"❌ No se pudo procesar imagen {i+j}: {str(e)}")
                 continue
-        
+
         # Guardar progreso
         if database:
             print("💾 Guardando progreso...")
             with open(EMBEDDINGS_FILE, 'wb') as f:
                 pickle.dump(database, f)
-        
+
         gc.collect()
-    
+
     return database
 
 # 🔍 Buscar rostros similares

metadata.py

@@ -1,40 +1,23 @@
 from huggingface_hub import HfApi
 import csv
 import os
-from pathlib import Path
 
-# 🔐 Configuración segura del token
-HF_TOKEN = os.getenv("HF_TOKEN")
-if not HF_TOKEN:
-    raise ValueError("⚠️ Por favor, configura la variable de entorno HF_TOKEN")
+HF_TOKEN = os.getenv("HF_TOKEN") or "hf_token"
+repo_id = "Segizu/facial-recognition"
 
-# 🗂️ Configurar repositorio
 api = HfApi()
-repo_id = "Segizu/facial-recognition"
+files = api.list_repo_files(repo_id=repo_id, repo_type="dataset", token=HF_TOKEN)
+
+# Generar URLs completas
+base_url = f"https://huggingface.co/datasets/{repo_id}/resolve/main/"
+image_urls = [base_url + f for f in files if f.lower().endswith(".jpg")]
 
-try:
-    # ✅ Listar todos los archivos del dataset
-    files = api.list_repo_files(repo_id=repo_id, repo_type="dataset", token=HF_TOKEN)
-    
-    # Filtrar imágenes .jpg
-    image_files = [f for f in files if f.lower().endswith(".jpg")]
-    
-    # Guardar metadata.csv
-    metadata_path = Path("metadata.csv")
-    with open(metadata_path, "w", newline="") as f:
-        writer = csv.writer(f)
-        writer.writerow(["image"])
-        for img in image_files:
-            writer.writerow([img])
-    
-    print(f"✅ metadata.csv generado con {len(image_files)} imágenes.")
+# Escribir nuevo metadata.csv
+with open("metadata.csv", "w", newline="") as f:
+    writer = csv.writer(f)
+    writer.writerow(["image"])
+    for url in image_urls:
+        writer.writerow([url])
 
-except Exception as e:
-    print(f"❌ Error: {str(e)}")
-    if "401" in str(e):
-        print("⚠️ Error de autenticación. Verifica que tu token de Hugging Face sea válido.")
-    elif "404" in str(e):
-        print("⚠️ No se encontró el repositorio. Verifica que el nombre del repositorio sea correcto.")
-    else:
-        print("⚠️ Ocurrió un error inesperado.")
+print(f"✅ metadata.csv regenerado con URLs absolutas ({len(image_urls)} imágenes)")