modify steps

app_base.py

@@ -31,13 +31,16 @@ def create_demo() -> gr.Blocks:
         start_step: int,
         guidance_scale: float,
         generate_size: int,
-        adapter_weights: float,
-        enhance_face: bool = True,
+        pre_enhance: bool = True,
     ):
         w2 = 1.0
         run_task_time = 0
         time_cost_str = ''
         run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
+        if pre_enhance:
+            input_image = enhance_image(input_image, enhance_face=True)
+            input_image = input_image.resize((generate_size, generate_size))
+        run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
         run_model = base_run
         res_image = run_model(
             input_image,
@@ -50,7 +53,6 @@ def create_demo() -> gr.Blocks:
             num_steps,
             start_step,
             guidance_scale,
-            adapter_weights,
         )
         run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
         enhanced_image = enhance_image(res_image, enhance_face)
@@ -81,11 +83,10 @@ def create_demo() -> gr.Blocks:
                 start_step = gr.Slider(minimum=1, maximum=100, value=15, step=1, label="Start Step")
                 with gr.Accordion("Advanced Options", open=False):
                     guidance_scale = gr.Slider(minimum=0, maximum=20, value=0, step=0.5, label="Guidance Scale")
-                    generate_size = gr.Number(label="Generate Size", value=1024)
+                    generate_size = gr.Number(label="Generate Size", value=512)
                     mask_expansion = gr.Number(label="Mask Expansion", value=50, visible=True)
                     mask_dilation = gr.Slider(minimum=0, maximum=10, value=2, step=1, label="Mask Dilation")
-                    enhance_face = gr.Checkbox(label="Enhance Face", value=False)
-                    adapter_weights = gr.Slider(minimum=0, maximum=1, value=0.5, step=0.1, label="Adapter Weights", visible=False)
+                    pre_enhance = gr.Checkbox(label="Pre Enhance", value=True)
             with gr.Column():
                 seed = gr.Number(label="Seed", value=8)
                 w1 = gr.Number(label="W1", value=2)
@@ -109,7 +110,7 @@ def create_demo() -> gr.Blocks:
             outputs=[origin_area_image, croper],
         ).success(
             fn=image_to_image,
-            inputs=[origin_area_image, input_image_prompt, edit_prompt,seed,w1, num_steps, start_step, guidance_scale, generate_size, adapter_weights, enhance_face],
+            inputs=[origin_area_image, input_image_prompt, edit_prompt,seed,w1, num_steps, start_step, guidance_scale, generate_size, pre_enhance],
             outputs=[enhanced_image, generated_image, generated_cost],
         ).success(
             fn=restore_result,

enhance_utils.py

@@ -21,12 +21,14 @@ model_path = 'realesr-general-x4v3.pth'
 half = True if torch.cuda.is_available() else False
 upsampler = RealESRGANer(scale=4, model_path=model_path, model=model, tile=0, tile_pad=10, pre_pad=0, half=half)
 
-face_enhancer = GFPGANer(model_path='GFPGANv1.4.pth', upscale=1, arch='clean', channel_multiplier=2)
+face_enhancer = GFPGANer(model_path='GFPGANv1.4.pth', upscale=2, arch='clean', channel_multiplier=2)
 
 def enhance_image(
     pil_image: Image,
     enhance_face: bool = True,
+    scale: int = 2,
 ):
+    face_enhancer.upscale = scale
     img = cv2.cvtColor(np.array(pil_image), cv2.COLOR_RGB2BGR)
 
     h, w = img.shape[0:2]

inversion_run_base.py

@@ -79,7 +79,6 @@ def run(
     num_steps:int,
     start_step:int,
     guidance_scale:float,
-    adapter_weights:float,
 ):
     generator = torch.Generator().manual_seed(seed)