22

app.py

@@ -48,69 +48,31 @@ def prepare(prompt, is_polish_prompt):
 def inference(
     prompt,
     negative_prompt,
-    input_image,
-    image_scale=1.0,
-    control_mode='Canny',
-    control_context_scale = 0.75,
     seed=42,
     randomize_seed=True,
     guidance_scale=1.5,
     num_inference_steps=8,
     progress=gr.Progress(track_tqdm=True),
 ):
-    # timestamp = time.time()
-    # print(f"timestamp: {timestamp}")
-
-    # # process image
-    # print("DEBUG: process image")
-    # if input_image is None:
-    #     print("Error: input_image is empty.")
-    #     return None
-    
-    # # input_image, width, height = scale_image(input_image, image_scale)
-    # # control_mode='HED'
-    # processor_id = 'canny'
-    # if control_mode == 'HED':
-    #     processor_id = 'softedge_hed'
-    # if control_mode =='Depth':
-    #     processor_id = 'depth_midas'
-    # if control_mode =='MLSD':
-    #     processor_id = 'mlsd'
-    # if control_mode =='Pose':
-    #     processor_id = 'openpose_full'
-
-    # print(f"DEBUG: processor_id={processor_id}")
-    # processor = Processor(processor_id)
-
-    # # Width must be divisible by 16
-    # control_image, width, height = rescale_image(input_image, image_scale, 16)
-    # control_image = control_image.resize((1024, 1024))
-
-    # print("DEBUG: processor running")
-    # control_image = processor(control_image, to_pil=True)
-    # control_image = control_image.resize((width, height))
-
-    # print("DEBUG: control_image_torch")
-    # control_image_torch = get_image_latent(control_image, sample_size=[height, width])[:, :, 0]
-
-    # # generation
-    # if randomize_seed: seed = random.randint(0, MAX_SEED)
-    # generator = torch.Generator().manual_seed(seed)
-
-    # image = pipe(
-    #     prompt=prompt,
-    #     negative_prompt = negative_prompt,
-    #     height=height,
-    #     width=width,
-    #     generator=generator,
-    #     guidance_scale=guidance_scale,
-    #     control_image=control_image_torch,
-    #     num_inference_steps=num_inference_steps,
-    #     control_context_scale=control_context_scale,
-    # ).images[0]
-
-    # return image, seed, control_image
-    return True
+    timestamp = time.time()
+    print(f"timestamp: {timestamp}")
+
+
+    # generation
+    if randomize_seed: seed = random.randint(0, MAX_SEED)
+    generator = torch.Generator().manual_seed(seed)
+
+    image = pipe(
+        prompt=prompt,
+        negative_prompt = negative_prompt,
+        height=1024,
+        width=1024,
+        generator=generator,
+        guidance_scale=guidance_scale,
+        num_inference_steps=num_inference_steps
+    ).images[0]
+
+    return image, seed
 
 
 def read_file(path: str) -> str:
@@ -229,10 +191,6 @@ with gr.Blocks() as demo:
         inputs=[
             polished_prompt,
             negative_prompt,
-            input_image,
-            image_scale,
-            control_mode,
-            control_context_scale,
             seed,
             randomize_seed,
             guidance_scale,