Update app.py

app.py

@@ -69,7 +69,7 @@ def infer(
     if controlnet_enabled and control_image:
         controlnet_model = ControlNetModel.from_pretrained(CONTROLNET_MODES.get(control_mode))
         if model_id == "SD1.5 + lora Unet TextEncoder": 
-            pipe=StableDiffusionControlNetPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5",controlnet=controlnet_model)
+            pipe=StableDiffusionControlNetPipeline.from_pretrained("stable-deeffusion-v1-5/stable-diffusion-v1-5",controlnet=controlnet_model)
             pipe.unet = PeftModel.from_pretrained(pipe.unet, "um235/vCat_v2", subfolder="unet")
             pipe.text_encoder = PeftModel.from_pretrained(pipe.text_encoder, "um235/vCat_v2", subfolder="text_encoder")
         elif model_id == "SD1.5 + lora Unet TextEncoder" or model_id == "SD1.5 + lora Unet":
@@ -109,6 +109,7 @@ def infer(
         controlnet_conditioning_scale=control_strength,
         ip_adapter_image=ip_adapter_image,
     ).images[0]
+    
     if d_bckg:
         image=remove(image)
         
@@ -158,12 +159,12 @@ with gr.Blocks(css=css) as demo:
                 minimum=0,
                 maximum=2,
                 step=0.05,
-                value=1,  
+                value=0.85,  
             )
         with gr.Row():
                 d_bckg=gr.Checkbox(label="Delete Background", value=False)
                 ddim_use=gr.Checkbox(label="Enable DDIMScheduler", value=False)
-                distill_vae=gr.Checkbox(label="Use tiny VAE with distill model", value=True)
+                distill_vae=gr.Checkbox(label="Use tiny VAE with distill model", value=False)
 
            # pipe.scheduler = DDIMScheduler.from_config(pipe.scheduler.config, rescale_betas_zero_snr=True)
         with gr.Row():
@@ -224,7 +225,7 @@ with gr.Blocks(css=css) as demo:
                 max_lines=1,
                 placeholder="Enter a negative prompt",
                 visible=True,
-                value="worst quality, normal quality, low quality, low res, blurry, distortion, text, watermark, logo, banner, extra digits, cropped, jpeg artifacts,"
+                value="worst quality,low quality, low res, blurry, distortion, jpeg artifacts, backround"
             )
 
             seed = gr.Slider(
@@ -232,10 +233,10 @@ with gr.Blocks(css=css) as demo:
                 minimum=0,
                 maximum=MAX_SEED,
                 step=1,
-                value=235,
+                value=750242712,
             )
 
-            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+            randomize_seed = gr.Checkbox(label="Randomize seed", value=False)
 
             with gr.Row():
                 width = gr.Slider(
@@ -260,7 +261,7 @@ with gr.Blocks(css=css) as demo:
                     minimum=0.0,
                     maximum=10.0,
                     step=0.1,
-                    value=7.0,  
+                    value=7.5,  
                 )
 
                 num_inference_steps = gr.Slider(