Update app.py

app.py

@@ -45,18 +45,13 @@ from torchvision.transforms import InterpolationMode
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")  
 
-# vae = AutoencoderKL.from_pretrained('.', subfolder='vae')
-# scheduler = DDIMScheduler.from_pretrained('.', subfolder='scheduler')
-# image_encoder = CLIPVisionModelWithProjection.from_pretrained('.', subfolder="image_encoder")
-# feature_extractor = CLIPImageProcessor.from_pretrained('.', subfolder="feature_extractor")
-
 stable_diffusion_repo_path = "stabilityai/stable-diffusion-2-1-unclip"
 vae = AutoencoderKL.from_pretrained(stable_diffusion_repo_path, subfolder='vae')
 scheduler = DDIMScheduler.from_pretrained(stable_diffusion_repo_path, subfolder='scheduler')
 sd_image_variations_diffusers_path = 'lambdalabs/sd-image-variations-diffusers'
 image_encoder = CLIPVisionModelWithProjection.from_pretrained(sd_image_variations_diffusers_path, subfolder="image_encoder")
 feature_extractor = CLIPImageProcessor.from_pretrained(sd_image_variations_diffusers_path, subfolder="feature_extractor")
-unet = UNet2DConditionModel.from_pretrained('.', subfolder="unet7000")
+unet = UNet2DConditionModel.from_pretrained('.', subfolder="unet")
 
 pipe = DepthNormalEstimationPipeline(vae=vae,
                             image_encoder=image_encoder,
@@ -161,7 +156,7 @@ def run_demo():
                          value=10,
                      )
                         ensemble_size = gr.Slider(
-                         label="Ensemble size (1 will be enough. More steps, higher accuracy)",
+                         label="Ensemble size (4 will be enough. More steps, higher accuracy)",
                          minimum=1,
                          maximum=15,
                          step=1,