debug

app.py

@@ -33,8 +33,8 @@ if torch.cuda.get_device_properties(0).major >= 8:
 
 FLORENCE_MODEL, FLORENCE_PROCESSOR = load_florence_model(device=DEVICE)
 SAM_IMAGE_MODEL = load_sam_image_model(device=DEVICE)
-FLUX_INPAINTING_PIPELINE = FluxInpaintPipeline.from_pretrained(
-    "black-forest-labs/FLUX.1-schnell", torch_dtype=torch.bfloat16).to(DEVICE)
+# FLUX_INPAINTING_PIPELINE = FluxInpaintPipeline.from_pretrained(
+#     "black-forest-labs/FLUX.1-schnell", torch_dtype=torch.bfloat16).to(DEVICE)
 
 
 def resize_image_dimensions(
@@ -63,7 +63,7 @@ def is_image_empty(image: Image.Image) -> bool:
     return all(pixel == 0 for pixel in pixels)
 
 
-@spaces.GPU(duration=150)
+@spaces.GPU()
 @torch.inference_mode()
 @torch.autocast(device_type="cuda", dtype=torch.bfloat16)
 def process(
@@ -128,23 +128,23 @@ def process(
     mask = mask.resize((width, height), Image.LANCZOS)
     mask = mask.filter(ImageFilter.GaussianBlur(radius=10))
 
-    # return image, mask
-
-    if randomize_seed_checkbox:
-        seed_slicer = random.randint(0, MAX_SEED)
-    generator = torch.Generator().manual_seed(seed_slicer)
-    result = FLUX_INPAINTING_PIPELINE(
-        prompt=inpainting_prompt_text,
-        image=image,
-        mask_image=mask,
-        width=width,
-        height=height,
-        strength=strength_slider,
-        generator=generator,
-        num_inference_steps=num_inference_steps_slider
-    ).images[0]
-    print('INFERENCE DONE')
-    return result, mask
+    return image, mask
+
+    # if randomize_seed_checkbox:
+    #     seed_slicer = random.randint(0, MAX_SEED)
+    # generator = torch.Generator().manual_seed(seed_slicer)
+    # result = FLUX_INPAINTING_PIPELINE(
+    #     prompt=inpainting_prompt_text,
+    #     image=image,
+    #     mask_image=mask,
+    #     width=width,
+    #     height=height,
+    #     strength=strength_slider,
+    #     generator=generator,
+    #     num_inference_steps=num_inference_steps_slider
+    # ).images[0]
+    # print('INFERENCE DONE')
+    # return result, mask
 
 
 with gr.Blocks() as demo: