Update app.py

app.py

@@ -32,7 +32,7 @@ models_loaded = False
 def load_models():
     global unet, tokenizer_one, tokenizer_two, noise_scheduler, text_encoder_one, text_encoder_two
     global image_encoder, vae, UNet_Encoder, parsing_model, openpose_model, pipe
-    global models_loaded  # Déclarer la variable globale ici
+    global models_loaded
     
     if not models_loaded:
         base_path = 'yisol/IDM-VTON'
@@ -47,7 +47,15 @@ def load_models():
         text_encoder_two = CLIPTextModelWithProjection.from_pretrained(base_path, subfolder="text_encoder_2", torch_dtype=torch.float16, force_download=False)
         image_encoder = CLIPVisionModelWithProjection.from_pretrained(base_path, subfolder="image_encoder", torch_dtype=torch.float16, force_download=False)
         vae = AutoencoderKL.from_pretrained(base_path, subfolder="vae", torch_dtype=torch.float16, force_download=False)
-        UNet_Encoder = UNet2DConditionModel_ref.from_pretrained(base_path, subfolder="unet_encoder", torch_dtype=torch.float16, force_download=False)
+        
+        # Set the correct encoder_hid_dim_type here
+        UNet_Encoder = UNet2DConditionModel.from_pretrained(
+            base_path,
+            subfolder="unet_encoder",
+            torch_dtype=torch.float16,
+            encoder_hid_dim_type="text_proj",  # Update based on model type
+            force_download=False
+        )
         
         parsing_model = Parsing(0)
         openpose_model = OpenPose(0)