Update app.py

app.py

@@ -12,69 +12,84 @@ import torch
 import numpy as np
 import cv2
 
-
-controlnet = ControlNetModel.from_pretrained(
-    "diffusers/controlnet-canny-sdxl-1.0",
-    torch_dtype=torch.float16
-)
 #vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
-pipe = StableDiffusionXLControlNetPipeline.from_pretrained(
-    "stabilityai/stable-diffusion-xl-base-1.0",
-    controlnet=controlnet,
-    #vae=vae,
-    torch_dtype=torch.float16, 
-    variant="fp16",
-    use_safetensors=True
-)
-pipe.to("cuda")
 
 generator = torch.Generator(device="cuda")
 
 #pipe.enable_model_cpu_offload()
 
-def infer(model_name, image_in, prompt, controlnet_conditioning_scale, guidance_scale, seed):
-    custom_model = model_name
+def infer(use_custom_model, model_name, image_in, prompt, preprocessor, controlnet_conditioning_scale, guidance_scale, seed):
+    if use_custom_model:
+        custom_model = model_name
 
-    # This is where you load your trained weights
-    pipe.load_lora_weights(custom_model, weight_name="pytorch_lora_weights.safetensors", use_auth_token=True)
+        # This is where you load your trained weights
+        pipe.load_lora_weights(custom_model, weight_name="pytorch_lora_weights.safetensors", use_auth_token=True)
     
     prompt = prompt
     negative_prompt = "extra digit, fewer digits, cropped, worst quality, low quality, glitch, deformed, mutated, ugly, disfigured"
 
-    image = load_image(image_in)
-
-    #controlnet_conditioning_scale = 0.25  # recommended for good generalization
-
-    image = np.array(image)
-    image = cv2.Canny(image, 100, 200)
-    image = image[:, :, None]
-    image = np.concatenate([image, image, image], axis=2)
-    image = Image.fromarray(image)
-
-    lora_scale= 0.9
-
-    
-    
-    images = pipe(
-        prompt, 
-        negative_prompt=negative_prompt, 
-        image=image, 
-        controlnet_conditioning_scale=controlnet_conditioning_scale,
-        guidance_scale = guidance_scale,
-        num_inference_steps=50,
-        generator=generator.manual_seed(seed),
-        cross_attention_kwargs={"scale": lora_scale}
+    if preprocessor == "canny":
+
+        controlnet = ControlNetModel.from_pretrained(
+            "diffusers/controlnet-canny-sdxl-1.0",
+            torch_dtype=torch.float16
+        )
+
+        pipe = StableDiffusionXLControlNetPipeline.from_pretrained(
+            "stabilityai/stable-diffusion-xl-base-1.0",
+            controlnet=controlnet,
+            #vae=vae,
+            torch_dtype=torch.float16, 
+            variant="fp16",
+            use_safetensors=True
+        )
+        pipe.to("cuda")
+        
+        image = load_image(image_in)
+
+        image = np.array(image)
+        image = cv2.Canny(image, 100, 200)
+        image = image[:, :, None]
+        image = np.concatenate([image, image, image], axis=2)
+        image = Image.fromarray(image)
+
+    if use_custom_model:
+        lora_scale= 0.9
+
+        images = pipe(
+            prompt, 
+            negative_prompt=negative_prompt, 
+            image=image, 
+            preprocessor=preprocessor,
+            controlnet_conditioning_scale=controlnet_conditioning_scale,
+            guidance_scale = guidance_scale,
+            num_inference_steps=50,
+            generator=generator.manual_seed(seed),
+            cross_attention_kwargs={"scale": lora_scale}
+        ).images
+    else:
+        images = pipe(
+            prompt, 
+            negative_prompt=negative_prompt, 
+            image=image, 
+            preprocessor=preprocessor,
+            controlnet_conditioning_scale=controlnet_conditioning_scale,
+            guidance_scale = guidance_scale,
+            num_inference_steps=50,
+            generator=generator.manual_seed(seed),
         ).images
 
-    images[0].save(f"hug_lab.png")
+    images[0].save(f"result.png")
 
-    return f"hug_lab.png"
+    return f"result.png"
 
 with gr.Blocks() as demo:
     with gr.Column():
+        use_custom_model = gr.Checkbox(label="Use a custom model ?", value=False)
         model_name = gr.Textbox(label="Model to use", placeholder="username/my_model")
         image_in = gr.Image(source="upload", type="filepath")
-        prompt = gr.Textbox(label="Prompt")
+        prompt = gr.Textbox(label="Prompt"),
+        preprocessor = gr.Dropdown(label="Preprocessor", choices=["canny"], value="canny")
         guidance_scale = gr.Slider(label="Guidance Scale", minimum=1.0, maximum=10.0, step=0.1, value=7.5, type="float")
         controlnet_conditioning_scale = gr.Slider(label="Controlnet conditioning Scale", minimum=0.1, maximum=0.9, step=0.01, value=0.5, type="float")
         seed = gr.Slider(label="seed", minimum=0, maximum=500000, step=1, value=42)
@@ -84,7 +99,7 @@ with gr.Blocks() as demo:
 
     submit_btn.click(
         fn = infer,
-        inputs = [model_name, image_in, prompt, controlnet_conditioning_scale, guidance_scale, seed],
+        inputs = [use_custom_model, model_name, image_in, prompt, preprocessor, controlnet_conditioning_scale, guidance_scale, seed],
         outputs = [result]
     )