【深度学习 AIGC】stablediffusion-infinity 在无界限画布中输出绘画 Outpainting
代码：https://github.com/lkwq007/stablediffusion-infinity/tree/master
启动环境：
git clone --recurse-submodules https://github.com/lkwq007/stablediffusion-infinity
cd stablediffusion-infinity
conda env create -f environment.yml
conda activate sd-inf
# 一定更新一下！
conda install -c conda-forge diffusers transformers ftfy accelerate
conda update -c conda-forge diffusers transformers ftfy accelerate
pip install -U gradio
python app.py
修改了一下app.py的东西，最后面修改了ip和端口：
launch_extra_kwargs = {
"show_error": True,
# "favicon_path": ""
}
launch_kwargs = vars(args)
launch_kwargs = {k: v for k, v in launch_kwargs.items() if v is not None}
print(launch_kwargs)
launch_kwargs.pop("remote_model", None)
launch_kwargs.pop("local_model", None)
launch_kwargs.pop("fp32", None)
launch_kwargs.pop("lowvram", None)
launch_kwargs.update(launch_extra_kwargs)
try:
import google.colab
launch_kwargs["debug"] = True
except:
pass
if RUN_IN_SPACE:
print("run in space")
demo.launch()
elif args.debug:
print(111111111)
launch_kwargs["share"]=True
launch_kwargs["server_name"] = "0.0.0.0"
launch_kwargs["server_port"] = 8000
demo.queue().launch(**launch_kwargs)
else:
print(222222222)
launch_kwargs["share"]=True
launch_kwargs["server_name"] = "0.0.0.0"
launch_kwargs["server_port"] = 8000
demo.queue().launch(**launch_kwargs)
可以对照一下环境：
(sd-inf) Thu Sep 14 20:59:37 /ssd/xiedong/stablediffusion-infinity pip list
Package Version
----------------------------- ---------
absl-py 1.3.0
accelerate 0.22.0
aiofiles 23.2.1
aiohttp 3.8.1
aiosignal 1.3.1
altair 5.1.1
antlr4-python3-runtime 4.9.3
anyio 3.6.2
async-timeout 4.0.2
attrs 23.1.0
backports.functools-lru-cache 1.6.4
bcrypt 4.0.1
brotlipy 0.7.0
cachetools 5.2.0
certifi 2023.7.22
cffi 1.15.1
charset-normalizer 2.0.4
click 8.1.3
cloudpickle 2.0.0
cmake 3.25.0
colorama 0.4.6
commonmark 0.9.1
contourpy 1.0.6
cryptography 38.0.1
cycler 0.11.0
cytoolz 0.12.0
dask 2022.7.0
dataclasses 0.8
datasets 2.7.0
diffusers 0.14.0
dill 0.3.6
einops 0.4.1
fastapi 0.87.0
ffmpy 0.3.0
filelock 3.8.0
fonttools 4.38.0
fpie 0.2.4
frozenlist 1.3.0
fsspec 2022.10.0
ftfy 6.1.1
google-auth 2.14.1
google-auth-oauthlib 0.4.6
gradio 3.44.2
gradio_client 0.5.0
grpcio 1.51.0
h11 0.12.0
httpcore 0.15.0
httpx 0.23.1
huggingface-hub 0.17.1
idna 3.4
imagecodecs 2021.8.26
imageio 2.19.3
importlib-metadata 5.0.0
importlib-resources 6.0.1
Jinja2 3.1.2
joblib 1.2.0
jsonschema 4.19.0
jsonschema-specifications 2023.7.1
kiwisolver 1.4.4
linkify-it-py 1.0.3
llvmlite 0.39.1
locket 1.0.0
Markdown 3.4.1
markdown-it-py 2.1.0
MarkupSafe 2.1.1
matplotlib 3.6.2
mdit-py-plugins 0.3.1
mdurl 0.1.2
mkl-fft 1.3.1
mkl-random 1.2.2
mkl-service 2.4.0
multidict 6.0.2
multiprocess 0.70.12.2
networkx 2.8.4
numba 0.56.4
numpy 1.23.4
oauthlib 3.2.2
omegaconf 2.2.3
opencv-python 4.6.0.66
opencv-python-headless 4.6.0.66
orjson 3.8.2
packaging 21.3
pandas 1.4.2
paramiko 2.12.0
partd 1.2.0
Pillow 9.2.0
pip 22.2.2
protobuf 3.20.3
psutil 5.9.1
pyarrow 8.0.0
pyasn1 0.4.8
pyasn1-modules 0.2.8
pycparser 2.21
pycryptodome 3.15.0
pydantic 1.10.2
pyDeprecate 0.3.2
pydub 0.25.1
Pygments 2.13.0
PyNaCl 1.5.0
pyOpenSSL 22.0.0
pyparsing 3.0.9
PySocks 1.7.1
python-dateutil 2.8.2
python-multipart 0.0.5
pytorch-lightning 1.7.7
pytz 2022.6
PyWavelets 1.3.0
PyYAML 6.0
referencing 0.30.2
regex 2022.4.24
requests 2.28.1
requests-oauthlib 1.3.1
responses 0.18.0
rfc3986 1.5.0
rich 12.6.0
rpds-py 0.10.3
rsa 4.9
sacremoses 0.0.53
safetensors 0.3.2
scikit-image 0.19.2
scipy 1.9.3
semantic-version 2.10.0
setuptools 65.5.0
six 1.16.0
sniffio 1.3.0
sourceinspect 0.0.4
starlette 0.21.0
taichi 1.2.2
tensorboard 2.11.0
tensorboard-data-server 0.6.1
tensorboard-plugin-wit 1.8.1
tifffile 2021.7.2
timm 0.6.11
tokenizers 0.11.4
toolz 0.12.0
torch 1.13.0
torchaudio 0.13.0
torchmetrics 0.10.3
torchvision 0.14.0
tqdm 4.64.1
transformers 4.33.1
typing_extensions 4.3.0
uc-micro-py 1.0.1
urllib3 1.26.12
uvicorn 0.20.0
wcwidth 0.2.5
websockets 10.4
Werkzeug 2.2.2
wheel 0.37.1
xxhash 0.0.0
yarl 1.7.2
zipp 3.10.0
路径下建立一个stabilityai，然后下载stable-diffusion-2-inpainting放进去，sd-vae-ft-mse是stable-diffusion-2-inpainting/vae里的东西复制了一遍。
(sd-inf) Thu Sep 14 21:00:31 /ssd/xiedong/stablediffusion-infinity tree stabilityai/
stabilityai/
├── sd-vae-ft-mse
│ ├── config.json
│ ├── diffusion_pytorch_model.bin
│ ├── diffusion_pytorch_model.fp16.bin
│ ├── diffusion_pytorch_model.fp16.safetensors
│ └── diffusion_pytorch_model.safetensors
└── stable-diffusion-2-inpainting
├── 512-inpainting-ema.ckpt
├── 512-inpainting-ema.safetensors
├── feature_extractor
│ └── preprocessor_config.json
├── merged-leopards.png
├── model_index.json
├── README.md
├── scheduler
│ └── scheduler_config.json
├── sd-vae-ft-mse-original
│ ├── README.md
│ ├── vae-ft-mse-840000-ema-pruned.ckpt
│ └── vae-ft-mse-840000-ema-pruned.safetensors
├── text_encoder
│ ├── config.json
│ ├── model.fp16.safetensors
│ ├── model.safetensors
│ ├── pytorch_model.bin
│ └── pytorch_model.fp16.bin
├── tokenizer
│ ├── merges.txt
│ ├── special_tokens_map.json
│ ├── tokenizer_config.json
│ └── vocab.json
├── unet
│ ├── config.json
│ ├── diffusion_pytorch_model.bin
│ ├── diffusion_pytorch_model.fp16.bin
│ ├── diffusion_pytorch_model.fp16.safetensors
│ └── diffusion_pytorch_model.safetensors
└── vae
├── config.json
├── diffusion_pytorch_model.bin
├── diffusion_pytorch_model.fp16.bin
├── diffusion_pytorch_model.fp16.safetensors
└── diffusion_pytorch_model.safetensors
然后就可以用了：
