Update app.py

app.py

@@ -1,3 +1,6 @@
+# ======================
+# --- 0. LIBRERIAS ---
+# ======================
 import gradio as gr
 import torch
 import torch.nn as nn
@@ -6,11 +9,15 @@ import torchvision.models as models
 import torchvision.transforms as transforms
 from PIL import Image
 
+# ========================
 # --- 1. CONFIGURACIÓN ---
+# ========================
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-imsize = 192 # Balance perfecto entre velocidad (CPU) y calidad
+imsize = 256 # MODIFICAR TAMAÑO DE IMAGEN (384,192,256) SI USAS VERSIONES DE SPACE GRATUITAS PARA NO IR LENTO :P
+
+
+# Transformación de entrada 
 
-# Transformación de entrada (Ahora redimensiona todo a imsize x imsize)
 loader = transforms.Compose([
     transforms.Resize((imsize, imsize)),
     transforms.ToTensor(),
@@ -18,6 +25,7 @@ loader = transforms.Compose([
 ])
 
 # Transformación inversa (Desnormalizar para mostrar la imagen final)
+
 unloader = transforms.Compose([
     transforms.Normalize(mean=[-0.485/0.229, -0.456/0.224, -0.406/0.225], 
                          std=[1/0.229, 1/0.224, 1/0.225]),
@@ -25,7 +33,9 @@ unloader = transforms.Compose([
     transforms.ToPILImage()
 ])
 
+# ===============================
 # --- 2. FUNCIONES DE PÉRDIDA ---
+# ===============================
 def calc_content_loss(gen_features, content_features):
     return torch.mean((gen_features - content_features) ** 2)
 
@@ -44,7 +54,9 @@ def calc_tv_loss(img):
     tv_w = torch.sum((img[:, :, :, 1:] - img[:, :, :, :-1]) ** 2)
     return tv_h + tv_w
 
+# ============================
 # --- 3. MODELO EXTRACTOR ---
+# ============================
 class VGGFeatureExtractor(nn.Module):
     def __init__(self):
         super().__init__()
@@ -63,8 +75,10 @@ class VGGFeatureExtractor(nn.Module):
             if name in self.style_layers: style_features[self.style_layers[name]] = x
             if name in self.content_layers: content_features[self.content_layers[name]] = x
         return content_features, style_features
-
+        
+# ========================================
 # --- 4. FUNCIÓN PRINCIPAL PARA GRADIO ---
+# ========================================
 def run_style_transfer(content_img, style_img, content_weight, style_weight, tv_weight, iterations):
     if content_img is None or style_img is None:
         return None
@@ -108,7 +122,10 @@ def run_style_transfer(content_img, style_img, content_weight, style_weight, tv_
     final_image = unloader(gen_img.cpu().squeeze(0))
     return final_image
 
+# =======================================
 # --- 5. INTERFAZ DE USUARIO (GRADIO) ---
+# =======================================
+
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     
     # ENCABEZADO Y ENLACES