Update app.py

app.py

@@ -59,7 +59,7 @@ refiner.to("cuda")
 MAX_SEED = np.iinfo(np.int32).max
 MAX_IMAGE_SIZE = 1024
 
-def ourhood_inference(prompt=str,num_inference_steps=int,scaffold=int,fracc=float):
+def ourhood_inference(prompt=str,num_inference_steps=int,scaffold=int,seed):
 
 ###pro_encode = pipe_cn.encode_text(prompt)
 
@@ -70,10 +70,10 @@ def ourhood_inference(prompt=str,num_inference_steps=int,scaffold=int,fracc=floa
                   'canny_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/mask_depth_solo_square.png"},
                2:{'mask1':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/mask_in_C.png",
                   'depth_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/depth_C.png",
-                  'canny_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/canny_C.png"},
+                  'canny_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/canny_C_solo.png"},
                3:{'mask1':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/mask_in_B.png",
                   'depth_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/depth_B.png",
-                  'canny_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/canny_B.png"}}
+                  'canny_image':"https://huggingface.co/Tonioesparza/ourhood_training_dreambooth_lora_2_0/resolve/main/canny_B_solo.png"}}
 ### mask init
 
     output_height = 1024
@@ -100,26 +100,34 @@ def ourhood_inference(prompt=str,num_inference_steps=int,scaffold=int,fracc=floa
 
     images_CN = [depth_image, canny_image]
 
+    
+
 ### inference
 
+    generator = torch.Generator(device="cuda").manual_seed(seed)
+
     results = pipe_CN(
         prompt=prompt,
         ip_adapter_image=ip_images,
         negative_prompt="deformed, ugly, wrong proportion, low res, worst quality, low quality,text,watermark",
+        generator=generator,
         num_inference_steps=n_steps,
         num_images_per_prompt=1,
-        denoising_end=fracc,
+        denoising_end=0.95,
         image=images_CN,
         output_type="latent",
-        controlnet_conditioning_scale=[0.3, 0.45],
+        control_guidance_start=[0.0, 0.35],
+        control_guidance_end=[0.35, 1.0],
+        controlnet_conditioning_scale=[0.5, 1.0],
         cross_attention_kwargs={"ip_adapter_masks": masks}
     ).images[0]
 
 
     image = refiner(
         prompt=prompt,
+        generator=generator,
         num_inference_steps=num_inference_steps,
-        denoising_start=fracc,
+        denoising_start=0.95,
         image=results,
     ).images[0]
 
@@ -184,13 +192,6 @@ with gr.Blocks(css=css) as demo:
 
             with gr.Row():
                 
-                fracc = gr.Slider(
-                    label="refinement_scale",
-                    minimum=0.8,
-                    maximum=0.95,
-                    step=0.01,
-                    value=0.8, #Replace with defaults that work for your model
-                )
                 
                 num_inference_steps = gr.Slider(
                     label="Number of inference steps",
@@ -207,7 +208,7 @@ with gr.Blocks(css=css) as demo:
     gr.on(
         triggers=[run_button.click, prompt.submit],
         fn = ourhood_inference,
-        inputs = [prompt, num_inference_steps, perspective, fracc],
+        inputs = [prompt, num_inference_steps, perspective, seed],
         outputs = [result]
     )