upload

app.py

@@ -62,7 +62,6 @@ def process_image(
     path_input,
     processing_res=default_image_processing_res,
 ):
-    assert False
     name_base, name_ext = os.path.splitext(os.path.basename(path_input))
     print(f"Processing image {name_base}{name_ext}")
 
@@ -267,9 +266,10 @@ def main():
 
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     
-    vae = AutoencoderKL.from_pretrained("guangkaixu/GenPercept", subfolder='vae')
-    unet_depth_v1 = UNet2DConditionModel.from_pretrained('guangkaixu/GenPercept', subfolder="unet_depth_v1")
-    dtype = unet_depth_v1.dtype
+    dtype = torch.float16
+    
+    vae = AutoencoderKL.from_pretrained("guangkaixu/GenPercept", subfolder='vae').to(dtype)
+    unet_depth_v1 = UNet2DConditionModel.from_pretrained('guangkaixu/GenPercept', subfolder="unet_depth_v1").to(dtype)
     empty_text_embed = torch.from_numpy(np.load("./empty_text_embed.npy")).to(device, dtype)[None] # [1, 77, 1024]
     
     pipe = GenPerceptPipeline(vae=vae,