ling

ling

fooocus_version.py

@@ -1 +1 @@
-version = '2.1.703'
+version = '2.1.704'

modules/async_worker.py

@@ -400,8 +400,9 @@ def worker():
 
             pipeline.final_unet.model.diffusion_model.in_inpaint = True
 
-            # outputs.append(['results', inpaint_worker.current_task.visualize_mask_processing()])
-            # return
+            if advanced_parameters.debugging_cn_preprocessor:
+                outputs.append(['results', inpaint_worker.current_task.visualize_mask_processing()])
+                return
 
             progressbar(13, 'VAE Inpaint encoding ...')
 

modules/default_pipeline.py

@@ -335,8 +335,6 @@ def process_diffusion(positive_cond, negative_cond, steps, switch, width, height
         # Disable refiner to avoid SD15 in joint swap
         final_refiner_unet = None
 
-    print(f'[Sampler] refiner_swap_method = {refiner_swap_method}')
-
     if latent is None:
         empty_latent = core.generate_empty_latent(width=width, height=height, batch_size=1)
     else:
@@ -353,6 +351,12 @@ def process_diffusion(positive_cond, negative_cond, steps, switch, width, height
         sigma_min, sigma_max, seed=image_seed, cpu=False)
 
     decoded_latent = None
+    refiner_use_different_vae = final_refiner_vae is not None and final_refiner_unet is not None
+
+    if refiner_swap_method == 'upscale' and not refiner_use_different_vae:
+        refiner_swap_method = 'joint'
+
+    print(f'[Sampler] refiner_swap_method = {refiner_swap_method}')
 
     if refiner_swap_method == 'joint':
         sampled_latent = core.ksampler(
@@ -375,14 +379,10 @@ def process_diffusion(positive_cond, negative_cond, steps, switch, width, height
         decoded_latent = core.decode_vae(vae=final_vae, latent_image=sampled_latent, tiled=tiled)
 
     if refiner_swap_method == 'upscale':
-        target_model = final_refiner_unet
-        if target_model is None:
-            target_model = final_unet
-
         sampled_latent = core.ksampler(
-            model=target_model,
-            positive=clip_separate(positive_cond, target_model=target_model.model, target_clip=final_clip),
-            negative=clip_separate(negative_cond, target_model=target_model.model, target_clip=final_clip),
+            model=final_refiner_unet,
+            positive=clip_separate(positive_cond, target_model=final_refiner_unet.model, target_clip=final_clip),
+            negative=clip_separate(negative_cond, target_model=final_refiner_unet.model, target_clip=final_clip),
             latent=empty_latent,
             steps=steps, start_step=0, last_step=steps, disable_noise=False, force_full_denoise=True,
             seed=image_seed,
@@ -394,11 +394,7 @@ def process_diffusion(positive_cond, negative_cond, steps, switch, width, height
             previewer_start=0,
             previewer_end=steps,
         )
-
-        target_model = final_refiner_vae
-        if target_model is None:
-            target_model = final_vae
-        decoded_latent = core.decode_vae(vae=target_model, latent_image=sampled_latent, tiled=tiled)
+        decoded_latent = core.decode_vae(vae=final_refiner_vae, latent_image=sampled_latent, tiled=tiled)
 
     if refiner_swap_method == 'separate':
         sampled_latent = core.ksampler(

modules/path.py

@@ -189,6 +189,15 @@ def downloading_inpaint_models(v):
         model_dir=inpaint_models_path,
         file_name='fooocus_inpaint_head.pth'
     )
+    head_file = os.path.join(inpaint_models_path, 'fooocus_inpaint_head.pth')
+    patch_file = None
+
+    # load_file_from_url(
+    #     url='https://huggingface.co/lllyasviel/Annotators/resolve/main/ControlNetLama.pth',
+    #     model_dir=inpaint_models_path,
+    #     file_name='ControlNetLama.pth'
+    # )
+    # lama_file = os.path.join(inpaint_models_path, 'ControlNetLama.pth')
 
     if v == 'v1':
         load_file_from_url(
@@ -196,8 +205,7 @@ def downloading_inpaint_models(v):
             model_dir=inpaint_models_path,
             file_name='inpaint.fooocus.patch'
         )
-        return os.path.join(inpaint_models_path, 'fooocus_inpaint_head.pth'), os.path.join(inpaint_models_path,
-                                                                                           'inpaint.fooocus.patch')
+        patch_file = os.path.join(inpaint_models_path, 'inpaint.fooocus.patch')
 
     if v == 'v2.5':
         load_file_from_url(
@@ -205,8 +213,9 @@ def downloading_inpaint_models(v):
             model_dir=inpaint_models_path,
             file_name='inpaint_v25.fooocus.patch'
         )
-        return os.path.join(inpaint_models_path, 'fooocus_inpaint_head.pth'), os.path.join(inpaint_models_path,
-                                                                                           'inpaint_v25.fooocus.patch')
+        patch_file = os.path.join(inpaint_models_path, 'inpaint_v25.fooocus.patch')
+
+    return head_file, patch_file
 
 
 def downloading_controlnet_canny():

webui.py

@@ -279,7 +279,7 @@ with shared.gradio_root:
                                                      info='Version of Fooocus inpaint model')
 
                     with gr.Tab(label='Control Debug'):
-                        debugging_cn_preprocessor = gr.Checkbox(label='Debug Preprocessor of ControlNets', value=False)
+                        debugging_cn_preprocessor = gr.Checkbox(label='Debug Preprocessors', value=False)
 
                         mixing_image_prompt_and_vary_upscale = gr.Checkbox(label='Mixing Image Prompt and Vary/Upscale',
                                                                            value=False)