Update app.py

app.py

@@ -1,7 +1,7 @@
 import spaces
 import gradio as gr
 import torch
-from diffusers import ControlNetModel, StableDiffusionXLControlNetImg2ImgPipeline, ControlNetModel, AutoencoderKL
+from diffusers import ControlNetModel, StableDiffusionXLControlNetImg2ImgPipeline, AutoencoderKL
 from PIL import Image
 import os
 import time
@@ -10,6 +10,7 @@ from utils.dl_utils import dl_cn_model, dl_cn_config, dl_lora_model
 from utils.image_utils import resize_image_aspect_ratio, base_generation
 from utils.prompt_utils import remove_duplicates
 
+# Setup directories and download necessary models
 path = os.getcwd()
 cn_dir = f"{path}/controlnet"
 lora_dir = f"{path}/lora"
@@ -20,6 +21,7 @@ dl_cn_model(cn_dir)
 dl_cn_config(cn_dir)
 dl_lora_model(lora_dir)
 
+# Model loading function
 def load_model(lora_dir, cn_dir):
     dtype = torch.float16
     vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
@@ -32,6 +34,7 @@ def load_model(lora_dir, cn_dir):
     pipe.load_lora_weights(lora_dir, weight_name="Fixhands_anime_bdsqlsz_V1.safetensors")
     return pipe
 
+# Image prediction and processing function
 @spaces.GPU(duration=120)
 def predict(input_image_path, prompt, negative_prompt, controlnet_scale):
     pipe = load_model(lora_dir, cn_dir) 
@@ -50,7 +53,7 @@ def predict(input_image_path, prompt, negative_prompt, controlnet_scale):
         control_image=resize_image,
         strength=1.0,
         prompt=prompt,
-        negative_prompt = negative_prompt,
+        negative_prompt=negative_prompt,
         controlnet_conditioning_scale=float(controlnet_scale),
         generator=generator,
         num_inference_steps=30,
@@ -67,7 +70,6 @@ class Img2Img:
         self.input_image_path = None
         self.canny_image = None
 
-
     def layout(self):
         css = """
         #intro{
@@ -78,14 +80,15 @@ class Img2Img:
         """
         with gr.Blocks(css=css) as demo:
             with gr.Row():
-                with gr.Column():
-                    self.input_image_path = gr.Image(label="input_image", type='filepath')
-                    self.prompt = gr.Textbox(label="prompt", lines=3)
-                    self.negative_prompt = gr.Textbox(label="negative_prompt", lines=3, value="nsfw, nipples, bad anatomy, liquid fingers, low quality, worst quality, out of focus, ugly, error, jpeg artifacts, lowers, blurry, bokeh")
-                    self.controlnet_scale = gr.Slider(minimum=0.5, maximum=2.0, value=1.0, step=0.01, label="Stick_fidelity")                 
-                    generate_button = gr.Button("generate")
-                with gr.Column():
-                    self.output_image = gr.Image(type="pil", label="output_image")
+                with gr.Column(scale=1):
+                    gr.Markdown("### Stickman to Posing Doll Image Converter\nこのアプリは棒人間をポーズ人形画像に変換するアプリです。\n入力する棒人間の形状は以下のリンクを参考にしてください。\n[VRoid Hub Character Example](https://hub.vroid.com/characters/4765753841994800453/models/6738034259079048708)\nIf your stick figure resembles the linked shape, it should work reasonably well even if hand-drawn.")
+                    self.input_image_path = gr.Image(label="Input Image", type='filepath')
+                    self.prompt = gr.Textbox(label="Prompt", lines=3)
+                    self.negative_prompt = gr.Textbox(label="Negative Prompt", lines=3, value="nsfw, nipples, bad anatomy, liquid fingers, low quality, worst quality, out of focus, ugly, error, jpeg artifacts, lowers, blurry, bokeh")
+                    self.controlnet_scale = gr.Slider(minimum=0.5, maximum=2.0, value=1.0, step=0.01, label="Controlnet Scale")
+                    generate_button = gr.Button("Generate")
+                with gr.Column(scale=1):
+                    self.output_image = gr.Image(type="pil", label="Output Image")
 
             generate_button.click(
                 fn=predict,