Update app.py

app.py

@@ -9,10 +9,7 @@ import torch
 import os
 from PIL import Image, ImageFilter, ImageOps
 from huggingface_hub import login, hf_hub_download
-import os
-
-# Add this import for theming
-from gradio import themes
+from gradio.themes import Default  # For theming
 
 
 if "HF_TOKEN" in os.environ:
@@ -49,7 +46,10 @@ if not os.path.exists(adetailer_model_path):
 adetailer_model = YOLO(adetailer_model_path)
 
 MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 1024 # @spaces.GPU #[uncomment to use ZeroGPU]
+MAX_IMAGE_SIZE = 1024
+
+# Purple theme
+custom_theme = Default(primary_hue="purple")
 
 def infer( 
     prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, scheduler_name="PNDM", save_format="png", progress=gr.Progress(track_tqdm=True),): 
@@ -70,16 +70,13 @@ def infer(
     # ---------------------------
     try:
         results = adetailer_model(image)
-        # ultralytics YOLO results typically have results[0].boxes
         if results and len(results) and getattr(results[0], "boxes", None):
             for box in results[0].boxes:
-                # extract bbox (xyxy)
                 x1, y1, x2, y2 = map(int, box.xyxy[0])
 
-                # 1) Pad bbox so inpaint has surrounding context
                 w = max(1, x2 - x1)
                 h = max(1, y2 - y1)
-                pad = int(max(10, 0.18 * max(w, h)))  # ~18% padding, tweak if desired to lower value for scenic panorama preservation
+                pad = int(max(10, 0.18 * max(w, h)))
                 x1p = max(0, x1 - pad)
                 y1p = max(0, y1 - pad)
                 x2p = min(image.width, x2 + pad)
@@ -87,45 +84,34 @@ def infer(
 
                 face = image.crop((x1p, y1p, x2p, y2p))
 
-                # 2) Make dimensions divisible by 8 (many pipelines prefer multiples of 8)
                 fw, fh = face.size
                 fw8 = max(8, (fw // 8) * 8)
                 fh8 = max(8, (fh // 8) * 8)
                 if (fw8, fh8) != (fw, fh):
                     face = face.resize((fw8, fh8), Image.LANCZOS)
 
-                # 3) Create a full white mask for inpainting (255 = area to replace)
                 mask = Image.new("L", face.size, 255)
 
-                # 4) Create a blurred mask for soft blending when pasting back
-                blur_radius = max(4, int(min(face.size) / 10))  # heuristic; adjust if needed
+                blur_radius = max(4, int(min(face.size) / 10))
                 paste_mask = mask.filter(ImageFilter.GaussianBlur(radius=blur_radius))
 
-                # 5) Inpaint on the crop — use gentler strength so output doesn't look pasted
-                #    (this re-uses your existing `pipe` call which previously worked for you)
                 inpaint_result = pipe(
                     prompt=prompt + ", high detail face",
                     image=face,
                     mask_image=mask,
-                    strength=0.45,                # reduced from 0.75 to lower mismatch
+                    strength=0.45,
                     num_inference_steps=20,
                     guidance_scale=7.5,
                     generator=generator
                 ).images[0]
 
-                # 6) Optionally attempt a simple color-match (blend) if lighting differs.
-                #    Here we keep it simple and rely on the blurred mask to smooth edges.
-                # Ensure paste_mask is 'L'
                 if paste_mask.mode != "L":
                     paste_mask = paste_mask.convert("L")
 
-                # 7) Paste the inpainted crop back using the soft mask for feathered edges
                 image.paste(inpaint_result, (x1p, y1p), paste_mask)
     except Exception as e:
-        # Log exception to console for debugging but don't crash the whole run
         print("ADetailer post-process failed:", e)
 
-    # Save in selected format for download consistency
     output_path = f"generated_image.{save_format}"
     image.save(output_path, format=save_format.upper())
     return image, seed 
@@ -135,14 +121,15 @@ examples = [
     "head helmet portrait of a futuristic armored soldier, worn brushed metal armor with neon blue accents, realistic cloth under-armor, weathering and scratches, volumetric rim light, cinematic pose, high detail, photoreal", 
     "close up view of silver coins on a table",] 
 
-css = """#col-container { margin: 0 auto; max-width: 640px;}""" 
-
-# Create the custom purple theme
-custom_theme = themes.Default(primary_hue="purple")
+# Updated CSS 12826
+css = """
+#col-container { margin: 0 auto; max-width: 640px;}
+#community-row {justify-content: center; gap: 30px;}
+"""
 
 with gr.Blocks(css=css, theme=custom_theme) as demo: 
     with gr.Column(elem_id="col-container"): 
-        gr.Markdown(" # DPS-Quagmaform AI txt2img") 
+        gr.Markdown("# DPS-Quagmaform AI txt2img") 
         with gr.Row(): 
             prompt = gr.Text( 
                 label="Prompt", show_label=False, max_lines=1, placeholder="Enter your prompt", container=False, 
@@ -181,6 +168,14 @@ with gr.Blocks(css=css, theme=custom_theme) as demo:
                 choices=["png", "jpg"], value="png", label="Select Output Format" 
             ) 
         gr.Examples(examples=examples, inputs=[prompt]) 
+        
+        # Community
+        gr.Markdown("### Community")
+        with gr.Row(elem_id="community-row"):
+            
+            gr.Button("Join Discord 💬", link="https://discord.gg/deepspace", variant="primary")
+            gr.Button("Join Telegram 📱", link="https://t.me/DeepSpaceHispano", variant="primary")
+        
         gr.on( 
             triggers=[run_button.click, prompt.submit], 
             fn=infer,