Update app.py

app.py

@@ -8,26 +8,23 @@ from accelerate import Accelerator
 from transformers import pipeline
 from diffusers.utils import load_image
 from diffusers import KandinskyV22PriorPipeline, KandinskyV22ControlnetPipeline
+from gradio_client import Client
 
 accelerator = Accelerator(cpu=True)
-MAX_SEED = np.iinfo(np.int32).max
-depth_estimator = accelerator.prepare(pipeline("depth-estimation", model="Intel/dpt-hybrid-midas"))
 pipe_prior = accelerator.prepare(KandinskyV22PriorPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float32))
 pipe_prior = accelerator.prepare(pipe_prior.to("cpu"))
 pipe = accelerator.prepare(KandinskyV22ControlnetPipeline.from_pretrained("kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float32))
 pipe = accelerator.prepare(pipe.to("cpu"))
+generator = torch.Generator("cpu").manual_seed(random.randint(1, 4867346))
 
-generator = torch.Generator("cpu").manual_seed(random.randint(1, MAX_SEED))
-
-def make_hint(image, depth_estimator):
-    image = depth_estimator(image)['depth']
-    image = load_image(image)
-    image.save('./dpt.png', 'PNG')
+def make_hint(image):
+    client = Client("https://nielsr-dpt-depth-estimation.hf.space/")
+    image = client.predict(image,api_name="/predict")
     image = np.array(image)
     image = image[:, :, None]
     image = np.concatenate([image, image, image], axis=2)
     detected_map = torch.from_numpy(image).float() / 255.0
-    hint = detected_map.permute(1, 2, 0)
+    hint = detected_map.permute(2, 0, 1)
     return hint
 
 def plex(goof,prompt):
@@ -35,8 +32,7 @@ def plex(goof,prompt):
     goof = load_image(goof)
     goof = goof.convert("RGB")
     goof.save('./gf.png', 'PNG')
-
-    hint = make_hint(goof, depth_estimator).unsqueeze(0).to("cpu")
+    hint = make_hint(goof).unsqueeze(0).to("cpu")
     negative_prior_prompt = "lowres,text,bad quality,jpeg artifacts,ugly,bad face,extra fingers,blurry,bad anatomy,extra limbs,fused fingers,long neck,watermark,signature"
     image_emb, zero_image_emb = pipe_prior(prompt=prompt, negative_prompt=negative_prior_prompt, num_inference_steps=5,generator=generator).to_tuple()