# Thank AK. https://huggingface.co/spaces/akhaliq/cool-japan-diffusion-2-1-0/blob/main/app.py from diffusers import StableDiffusionPipeline, StableDiffusionImg2ImgPipeline, EulerAncestralDiscreteScheduler,StableDiffusionLatentUpscalePipeline from transformers import CLIPImageProcessor import gradio as gr import torch from PIL import Image import random import spaces import requests # 得られたIPアドレスを使用してリクエストを行う # response = requests.get('https://huggingface.co/') # print(response.text) model_id = 'aipicasso/cool-japan-diffusion-2-1-2' scheduler = EulerAncestralDiscreteScheduler.from_pretrained(model_id, subfolder="scheduler") feature_extractor = CLIPImageProcessor.from_pretrained(model_id) pipe = StableDiffusionPipeline.from_pretrained( model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, scheduler=scheduler) #pipe.enable_xformers_memory_efficient_attention() #pipe.enable_freeu(s1=0.9, s2=0.2, b1=1.4, b2=1.6) # b1: 1.4, b2: 1.6, s1: 0.9, s2: 0.2 pipe_i2i = StableDiffusionImg2ImgPipeline.from_pretrained( model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, scheduler=scheduler, requires_safety_checker=False, safety_checker=None, feature_extractor=feature_extractor ) #pipe_i2i.enable_xformers_memory_efficient_attention() upscaler = StableDiffusionLatentUpscalePipeline.from_pretrained("alfredplpl/x2-latent-upscaler-for-anime", torch_dtype=torch.float16) #upscaler.enable_xformers_memory_efficient_attention() if torch.cuda.is_available(): pipe = pipe.to("cuda") pipe_i2i = pipe_i2i.to("cuda") upscaler=upscaler.to("cuda") def error_str(error, title="Error"): return f"""#### {title} {error}""" if error else "" @spaces.GPU def inference(prompt, guidance, steps, image_size="Square", seed=0, img=None, strength=0.5, neg_prompt="", cool_japan_type="Anime", disable_auto_prompt_correction=False): generator = torch.Generator('cuda').manual_seed(seed) if seed != 0 else None prompt,neg_prompt=auto_prompt_correction(prompt,neg_prompt,cool_japan_type,disable_auto_prompt_correction) if(image_size=="Portrait"): height=1024 width=768 superreso=False #pipe.enable_attention_slicing() elif(image_size=="Landscape"): height=768 width=1024 superreso=False #pipe.enable_attention_slicing() elif(image_size=="Highreso."): height=1024 width=1024 superreso=False #pipe.enable_attention_slicing() elif(image_size=="Superreso."): height=1024 width=1024 superreso=True #pipe.enable_attention_slicing() else: height=768 width=768 superreso=False #pipe.enable_attention_slicing() print(prompt,neg_prompt) try: if img is not None: return img_to_img(prompt, neg_prompt, img, strength, guidance, steps, width, height, generator,superreso), None else: return txt_to_img(prompt, neg_prompt, guidance, steps, width, height, generator,superreso), None except Exception as e: return None, error_str(e) def auto_prompt_correction(prompt_ui,neg_prompt_ui,cool_japan_type_ui,disable_auto_prompt_correction): # auto prompt correction cool_japan_type=str(cool_japan_type_ui) if(cool_japan_type=="Manga"): cool_japan_type="manga, monochrome, white and black manga" elif(cool_japan_type=="Game"): cool_japan_type="game" else: cool_japan_type="anime" prompt=str(prompt_ui) neg_prompt=str(neg_prompt_ui) prompt=prompt.lower() neg_prompt=neg_prompt.lower() if(disable_auto_prompt_correction): prompt=f"{cool_japan_type}, {prompt}" return prompt, neg_prompt if(prompt=="" and neg_prompt==""): #prefix=["masterpiece","evangelion, mika pikazo", "konosuba, mika pikazo","steins; gate, ilya kuvshinov", # "ghibli, shinkai makoto", "evangelion, madoka magica"] #suffix=["","ayanami rei, asuka langrey", "megumin, aqua from konosuba","mayuri shiina from steins gate, kurisu makise steins gate anime", # "hakurei reimu","kirisame marisa", "kaname madoka, megumin"] #prefix_index=random.randrange(len(prefix)) #suffix_index=random.randrange(len(suffix)) #prompt=f"{cool_japan_type}, {prefix[prefix_index]}, portrait, a good girl, {suffix[suffix_index]}, good pupil, 4k, detailed" prompt=f"{cool_japan_type}, masterpiece, upper body, a girl, good pupil, 4k, detailed" neg_prompt=f"(((deformed))), blurry, ((((bad anatomy)))), bad pupil, disfigured, poorly drawn face, mutation, mutated, (extra limb), (ugly), (poorly drawn hands), bad hands, fused fingers, messy drawing, broken legs censor, low quality, (mutated hands and fingers:1.5), (long body :1.3), (mutation, poorly drawn :1.2), ((bad eyes)), ui, error, missing fingers, fused fingers, one hand with more than 5 fingers, one hand with less than 5 fingers, one hand with more than 5 digit, one hand with less than 5 digit, extra digit, fewer digits, fused digit, missing digit, bad digit, liquid digit, long body, uncoordinated body, unnatural body, lowres, jpeg artifacts, 3d, cg, text" return prompt, neg_prompt splited_prompt=prompt.replace(","," ").replace("_"," ").split(" ") splited_prompt=["a person" if p=="solo" else p for p in splited_prompt] splited_prompt=["girl" if p=="1girl" else p for p in splited_prompt] splited_prompt=["a couple of girls" if p=="2girls" else p for p in splited_prompt] splited_prompt=["a couple of boys" if p=="2boys" else p for p in splited_prompt] human_words=["girl","maid","maids","female","woman","girls","a couple of girls","women","boy","boys","a couple of boys","male","man","men","guy","guys"] for word in human_words: if( word in splited_prompt): prompt=f"{cool_japan_type}, masterpiece, {prompt}, good pupil, 4k, detailed" neg_prompt=f"(((deformed))), blurry, ((((bad anatomy)))), {neg_prompt}, bad pupil, disfigured, poorly drawn face, mutation, mutated, (extra limb), (ugly), (poorly drawn hands), bad hands, fused fingers, messy drawing, broken legs censor, low quality, (mutated hands and fingers:1.5), (long body :1.3), (mutation, poorly drawn :1.2), ((bad eyes)), ui, error, missing fingers, fused fingers, one hand with more than 5 fingers, one hand with less than 5 fingers, one hand with more than 5 digit, one hand with less than 5 digit, extra digit, fewer digits, fused digit, missing digit, bad digit, liquid digit, long body, uncoordinated body, unnatural body, lowres, jpeg artifacts, 3d, cg, text" animal_words=["cat","dog","bird","horse","pigeon"] for word in animal_words: if( word in splited_prompt): prompt=f"{cool_japan_type}, a {word}, 4k, detailed" neg_prompt=f"(((deformed))), blurry, ((((bad anatomy)))), {neg_prompt}, bad pupil, disfigured, poorly drawn face, mutation, mutated, (extra limb), (ugly), (poorly drawn hands), bad hands, fused fingers, messy drawing, broken legs censor, low quality, (mutated hands and fingers:1.5), (long body :1.3), (mutation, poorly drawn :1.2), ((bad eyes)), ui, error, missing fingers, fused fingers, one hand with more than 5 fingers, one hand with less than 5 fingers, one hand with more than 5 digit, one hand with less than 5 digit, extra digit, fewer digits, fused digit, missing digit, bad digit, liquid digit, long body, uncoordinated body, unnatural body, lowres, jpeg artifacts, 3d, cg, text" background_words=["mount fuji","mt. fuji","building", "buildings", "tokyo", "kyoto", "nara", "shibuya", "shinjuku"] for word in background_words: if( word in splited_prompt): prompt=f"{cool_japan_type}, shinkai makoto, {word}, 4k, 8k, highly detailed" neg_prompt=f"(((deformed))), {neg_prompt}, girl, boy, photo, people, low quality, ui, error, lowres, jpeg artifacts, 2d, 3d, cg, text" return prompt,neg_prompt def txt_to_img(prompt, neg_prompt, guidance, steps, width, height, generator,superreso=False): global pipe, upscaler if(superreso): low_res_latents = pipe( prompt, negative_prompt = neg_prompt, num_inference_steps = int(steps), guidance_scale = guidance, width = width, height = height, output_type="latent", generator = generator).images result = upscaler( prompt=prompt, negative_prompt = neg_prompt, image=low_res_latents, num_inference_steps=20, guidance_scale=0, generator=generator, ) else: result = pipe( prompt, negative_prompt = neg_prompt, num_inference_steps = int(steps), guidance_scale = guidance, width = width, height = height, generator = generator) return result.images[0] def img_to_img(prompt, neg_prompt, img, strength, guidance, steps, width, height, generator,superreso=False): ratio = min(height / img.height, width / img.width) img = img.resize((int(img.width * ratio), int(img.height * ratio)), Image.LANCZOS) if(superreso): low_res_latents = pipe_i2i( prompt, negative_prompt = neg_prompt, image = img, num_inference_steps = int(steps), strength = strength, guidance_scale = guidance, #width = width, #height = height, output_type="latent", generator = generator).images result = upscaler( prompt=prompt, negative_prompt = neg_prompt, image=low_res_latents, num_inference_steps=20, guidance_scale=0, generator=generator, ) else: result = pipe_i2i( prompt, negative_prompt = neg_prompt, image = img, num_inference_steps = int(steps), strength = strength, guidance_scale = guidance, #width = width, #height = height, generator = generator) return result.images[0] css = """.main-div div{display:inline-flex;align-items:center;gap:.8rem;font-size:1.75rem}.main-div div h1{font-weight:900;margin-bottom:7px}.main-div p{margin-bottom:10px;font-size:94%}a{text-decoration:underline}.tabs{margin-top:0;margin-bottom:0}#gallery{min-height:20rem} """ with gr.Blocks(css=css) as demo: gr.HTML( f"""

Cool Japan Diffusion 2.1.2

Demo for Cool Japan Diffusion 2.1.2 .

sample : Click "Generate" button without any prompts.

sample prompt1 : girl, kimono

sample prompt2 : boy, school uniform

Running on {"GPU 🔥" if torch.cuda.is_available() else f"CPU 🥶. For faster inference it is recommended to upgrade to GPU in Settings"}
Duplicate Space to say goodbye from waiting for the generating.
""" ) with gr.Row(): with gr.Column(scale=55): with gr.Group(): with gr.Row(): cool_japan_type=gr.Radio(["Anime", "Manga", "Game"]) cool_japan_type.show_label=False cool_japan_type.value="Anime" with gr.Row(): prompt = gr.Textbox(label="Prompt", show_label=False, max_lines=2,placeholder="[your prompt]") # .style(container=False) generate = gr.Button(value="Generate") # .style(rounded=(False, True, True, False)) image_out = gr.Image(height=768,width=768) error_output = gr.Markdown() with gr.Column(scale=45): with gr.Tab("Options"): with gr.Group(): neg_prompt = gr.Textbox(label="Negative prompt", placeholder="What to exclude from the image") disable_auto_prompt_correction = gr.Checkbox(label="Disable auto prompt corretion.") with gr.Row(): image_size=gr.Radio(["Portrait","Landscape","Square","Highreso.","Superreso."]) image_size.show_label=False image_size.value="Square" with gr.Row(): guidance = gr.Slider(label="Guidance scale", value=7.5, maximum=15) steps = gr.Slider(label="Steps", value=20, minimum=2, maximum=75, step=1) seed = gr.Slider(0, 2147483647, label='Seed (0 = random)', value=0, step=1) with gr.Tab("Image to image"): with gr.Group(): image = gr.Image(label="Image", height=256, sources="upload", type="pil") strength = gr.Slider(label="Transformation strength", minimum=0, maximum=1, step=0.01, value=0.5) inputs = [prompt, guidance, steps, image_size, seed, image, strength, neg_prompt, cool_japan_type, disable_auto_prompt_correction] outputs = [image_out, error_output] prompt.submit(inference, inputs=inputs, outputs=outputs) generate.click(inference, inputs=inputs, outputs=outputs,api_name="generate") demo.queue() demo.launch(share=True)