File size: 13,692 Bytes
5e70012 431f639 f811c83 c14628b a6ac31b 7f16cc8 78f9628 c14628b 8ed3b1a a4bb94e 2c7d2b1 1a17b1d f9756f1 0cd1825 c557830 c14628b a6ac31b 6303ab5 a6ac31b 6506b9e c557830 60d9680 a6ac31b 6303ab5 5e70012 6506b9e a6ac31b 305c78b 6506b9e 9d44341 431f639 a6ac31b 5810cdb a6ac31b 0731c3b 624e415 a6ac31b 5e70012 bfda42f 47f4d74 bfda42f 624e415 899c80f 431f639 8c44916 624e415 899c80f 431f639 8c44916 6303ab5 431f639 66f06a8 431f639 6303ab5 624e415 899c80f 431f639 8c44916 73627b7 a6ac31b 263cee5 431f639 263cee5 431f639 a6ac31b 47f4d74 5e70012 bfda42f 1db878b 5e70012 47f4d74 5e70012 f26498b de423d8 6d4d808 eb74a82 5e70012 a6c0409 7507930 42a9377 ee99142 5e70012 7d1ffa7 5434e08 5e70012 05ddcd1 5e70012 bfda42f 5434e08 5e70012 cd534b2 5e70012 bfda42f cd534b2 5e70012 78f9628 431f639 9d44341 431f639 b80f9f1 6c78f11 431f639 10289c4 32bc0ab 431f639 6c78f11 431f639 a6ac31b 431f639 a6ac31b 431f639 1a3ab60 431f639 b80f9f1 431f639 10289c4 32bc0ab 431f639 a6ac31b 1a3ab60 a6ac31b 5e70012 a6ac31b 431f639 a6ac31b fe3bbad a6ac31b fe3bbad a6ac31b 5e70012 899c80f 2db0441 578fb43 5e70012 578fb43 5e70012 8f1b415 09dd938 a6ac31b bfda42f 1db878b 7893837 a18414d eb9b478 a6ac31b 899c80f a6ac31b 5e70012 361d7a3 431f639 361d7a3 899c80f 361d7a3 a6ac31b 5e70012 a6ac31b d9ae386 a6ac31b 5e70012 624e415 a6ac31b 844032f a6ac31b 2c7d2b1 401f598 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 |
# 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"""
<div class="main-div">
<div>
<h1>Cool Japan Diffusion 2.1.2</h1>
</div>
<p>
Demo for <a href="https://huggingface.co/aipicasso/cool-japan-diffusion-2-1-2">Cool Japan Diffusion 2.1.2</a> .<br>
</p>
<p>
sample : Click "Generate" button without any prompts.
</p>
<p>
sample prompt1 : girl, kimono
</p>
<p>
sample prompt2 : boy, school uniform
</p>
Running on {"<b>GPU 🔥</b>" if torch.cuda.is_available() else f"<b>CPU 🥶</b>. For faster inference it is recommended to <b>upgrade to GPU in <a href='https://huggingface.co/spaces/akhaliq/cool-japan-diffusion-2-1-0/settings'>Settings</a></b>"} <br>
<a style="display:inline-block" href="https://huggingface.co/spaces/aipicasso/cool-japan-diffusion-latest-demo?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a> to say goodbye from waiting for the generating.
</div>
"""
)
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)
|