Skip to content

GitLab

"tests/git@developer.sourcefind.cn:OpenDAS/apex.git" did not exist on "13c8d1521a713a3f5e94c115b52ce39466ec07cc"
Commit 25008cf0 authored by dongchy920's avatar dongchy920
Browse files

test

parents
Pipeline #1767 canceled with stages
Hide whitespace changes
Inline Side-by-side
Showing with 1339 additions and 0 deletions
.gitignore 0 → 100644
View file @ 25008cf0
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class
# C extensions
*.so
# Distribution / packaging
.Python
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
pip-wheel-metadata/
share/python-wheels/
*.egg-info/
.installed.cfg
*.egg
MANIFEST
# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec
# Installer logs
pip-log.txt
pip-delete-this-directory.txt
# Unit test / coverage reports
htmlcov/
.tox/
.nox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
*.py,cover
.hypothesis/
.pytest_cache/
# Translations
*.mo
*.pot
# Django stuff:
*.log
local_settings.py
db.sqlite3
db.sqlite3-journal
# Flask stuff:
instance/
.webassets-cache
# Scrapy stuff:
.scrapy
# Sphinx documentation
docs/_build/
# PyBuilder
target/
# Jupyter Notebook
.ipynb_checkpoints
# IPython
profile_default/
ipython_config.py
# pyenv
.python-version
# pipenv
# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
# However, in case of collaboration, if having platform-specific dependencies or dependencies
# having no cross-platform support, pipenv may install dependencies that don't work, or not
# install all needed dependencies.
#Pipfile.lock
# PEP 582; used by e.g. github.com/David-OConnor/pyflow
__pypackages__/
# Celery stuff
celerybeat-schedule
celerybeat.pid
# SageMath parsed files
*.sage.py
# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/
# Spyder project settings
.spyderproject
.spyproject
# Rope project settings
.ropeproject
# mkdocs documentation
/site
# mypy
.mypy_cache/
.dmypy.json
dmypy.json
# Pyre type checker
.pyre/
wandb/
*.lmdb/
*.pkl
LICENSE 0 → 100644
View file @ 25008cf0
MIT License
Copyright (c) 2019 Kim Seonghyeon
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
LICENSE-FID 0 → 100644
View file @ 25008cf0
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
LICENSE-LPIPS 0 → 100644
View file @ 25008cf0
Copyright (c) 2018, Richard Zhang, Phillip Isola, Alexei A. Efros, Eli Shechtman, Oliver Wang
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
LICENSE-NVIDIA 0 → 100644
View file @ 25008cf0
Copyright (c) 2019, NVIDIA Corporation. All rights reserved.
Nvidia Source Code License-NC
=======================================================================
1. Definitions
"Licensor" means any person or entity that distributes its Work.
"Software" means the original work of authorship made available under
this License.
"Work" means the Software and any additions to or derivative works of
the Software that are made available under this License.
"Nvidia Processors" means any central processing unit (CPU), graphics
processing unit (GPU), field-programmable gate array (FPGA),
application-specific integrated circuit (ASIC) or any combination
thereof designed, made, sold, or provided by Nvidia or its affiliates.
The terms "reproduce," "reproduction," "derivative works," and
"distribution" have the meaning as provided under U.S. copyright law;
provided, however, that for the purposes of this License, derivative
works shall not include works that remain separable from, or merely
link (or bind by name) to the interfaces of, the Work.
Works, including the Software, are "made available" under this License
by including in or with the Work either (a) a copyright notice
referencing the applicability of this License to the Work, or (b) a
copy of this License.
2. License Grants
2.1 Copyright Grant. Subject to the terms and conditions of this
License, each Licensor grants to you a perpetual, worldwide,
non-exclusive, royalty-free, copyright license to reproduce,
prepare derivative works of, publicly display, publicly perform,
sublicense and distribute its Work and any resulting derivative
works in any form.
3. Limitations
3.1 Redistribution. You may reproduce or distribute the Work only
if (a) you do so under this License, (b) you include a complete
copy of this License with your distribution, and (c) you retain
without modification any copyright, patent, trademark, or
attribution notices that are present in the Work.
3.2 Derivative Works. You may specify that additional or different
terms apply to the use, reproduction, and distribution of your
derivative works of the Work ("Your Terms") only if (a) Your Terms
provide that the use limitation in Section 3.3 applies to your
derivative works, and (b) you identify the specific derivative
works that are subject to Your Terms. Notwithstanding Your Terms,
this License (including the redistribution requirements in Section
3.1) will continue to apply to the Work itself.
3.3 Use Limitation. The Work and any derivative works thereof only
may be used or intended for use non-commercially. The Work or
derivative works thereof may be used or intended for use by Nvidia
or its affiliates commercially or non-commercially. As used herein,
"non-commercially" means for research or evaluation purposes only.
3.4 Patent Claims. If you bring or threaten to bring a patent claim
against any Licensor (including any claim, cross-claim or
counterclaim in a lawsuit) to enforce any patents that you allege
are infringed by any Work, then your rights under this License from
such Licensor (including the grants in Sections 2.1 and 2.2) will
terminate immediately.
3.5 Trademarks. This License does not grant any rights to use any
Licensor's or its affiliates' names, logos, or trademarks, except
as necessary to reproduce the notices described in this License.
3.6 Termination. If you violate any term of this License, then your
rights under this License (including the grants in Sections 2.1 and
2.2) will terminate immediately.
4. Disclaimer of Warranty.
THE WORK IS PROVIDED "AS IS" WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE OR
NON-INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY ACTIVITIES UNDER
THIS LICENSE.
5. Limitation of Liability.
EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL
THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE
SHALL ANY LICENSOR BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT,
INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF
OR RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK
(INCLUDING BUT NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION,
LOST PROFITS OR DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER
COMMERCIAL DAMAGES OR LOSSES), EVEN IF THE LICENSOR HAS BEEN ADVISED OF
THE POSSIBILITY OF SUCH DAMAGES.
=======================================================================
README.md 0 → 100644
View file @ 25008cf0
# StyleGAN2
## 论文
- https://arxiv.org/pdf/1912.04958
## 模型结构
针对StyleGAN中，大多数的生成的图像容易产生一个类似水滴状的伪影问题，StyleGAN2对AdaIN的归一化操作进行改进，作者把AdaIN层里面的归一化去掉，将噪声B和偏置项b移动出style模块之外，取得了更好的生成效果。
<div align=center>
<img src="./assets/AdaIN.png"/>
</div>
## 算法原理
StyleGAN2为StyleGAN的改进版本，StyleGAN是在PGGAN的基础上进行改进的模型
GAN（生成对抗网络）包含了一个生成模型G和一个判别模型D，模型通过生成器G对从正态分布P(z)随机采样的z生成伪数据x'，和从真实图像分布P(data)中采样的样本x作为判别器D的输入，判别器要让x的概率越大越好，让x'的概率越小越好，同时生成器希望生成的样本让判别器判别为真的概率越大越好。通过这种对抗的方式使模型生成越来越逼真的图片。
<div align=center>
<img src="./assets/GAN2.png"/>
</div>
PGGAN（渐进式生长生成对抗网络）通过先从低分辨率开始训练，然后再逐层提高分辨率进行训练的方式解决了传统GAN存在的模式崩溃（生成数据只是原始数据的子集（生成器偏向于生成判别器难以判别的样本））和难以训练高分辨率图片（生成器刚开始直接生成高分辨率图片很容易被判别器识别，在反向传播出现梯度大范围更新导致生成器崩溃）的问题，PGGAN先训练低分辨率，然后通过平滑接入的方式逐步提高分辨率：
<div align=center>
<img src="./assets/PGGAN.png"/>
</div>
PGGAN虽然能生成高清伪图，但是不能对图象的风格和细节进行修改，StyleGAN通过对特征进行解耦，让特征之间相互独立，互不影响，从而达到单独修改图象的某一部分的目的。具体来说StyleGAN是通过修改生成器，下图中左边为传统的生成器，右部分为StyleGAN的生成器，由两部分构成——Mapping network和Synthesis network ，其中Mapping network就是用来控制图像的风格信息，Synthesis network用来生成图像
<div align=center>
<img src="./assets/styleGAN.png"/>
</div>
StyleGAN2就是在StyleGAN的基础上改进了归一化操作，对损失函数和训练方法进行改进。
## 环境配置
### Docker（方法一）
从[光源](https://www.sourcefind.cn/#/service-list)中拉取docker镜像：
```
docker pull image.sourcefind.cn:5000/dcu/admin/base/pytorch:2.1.0-ubuntu20.04-dtk24.04.2-py3.10
```
创建容器并挂载目录进行开发：
```
docker run -it --name {name} --shm-size=1024G --device=/dev/kfd --device=/dev/dri/ --privileged --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ulimit memlock=-1:-1 --ipc=host --network host --group-add video -v /opt/hyhal:/opt/hyhal:ro -v {}:{} {docker_image} /bin/bash
# 修改1 {name} 需要改为自定义名称，建议命名{框架_dtk版本_使用者姓名},如果有特殊用途可在命名框架前添加命名
# 修改2 {docker_image} 需要需要创建容器的对应镜像名称，如: image.sourcefind.cn:5000/dcu/admin/base/pytorch:2.1.0-ubuntu20.04-dtk24.04.2-py3.10
# 修改3 -v 挂载路径到容器指定路径
pip install -r requirements.txt
```
### Dockerfile（方法二）
```
cd docker
docker build --no-cache -t gan2_pytorch:1.0 .
docker run -it --name {name} --shm-size=1024G --device=/dev/kfd --device=/dev/dri/ --privileged --cap-add=SYS_PTRACE --security-opt seccomp=unconfined --ulimit memlock=-1:-1 --ipc=host --network host --group-add video -v /opt/hyhal:/opt/hyhal:ro -v {}:{} {docker_image} /bin/bash
pip install -r requirements.txt
```
### Anaconda（方法三）
线上节点推荐使用conda进行环境配置。
创建python=3.10的conda环境并激活
```
conda create -n styleGan2 python=3.10
conda activate styleGan2
```
关于本项目DCU显卡所需的特殊深度学习库可从[光合](https://developer.hpccube.com/tool/)开发者社区下载安装。
```
DTK驱动：dtk24.04.2
python：python3.10
pytorch:2.1.0
torchvision:0.16.0
```
安装其他依赖包
```
pip install -r requirements.txt
```
## 数据集
- unconditional models训练集[FFHQ](https://drive.google.com/drive/folders/1u2xu7bSrWxrbUxk-dT-UvEJq8IjdmNTP)，
- 用于快速复现的小数据集SCNet快速下载链接 [images](http://113.200.138.88:18080/aidatasets/project-dependency/stylegan2_pytorch)
下载后解压到当前目录
数据目录结构如下：
```
./data/ffhq/images
   └── 00000.png
└── 00001.png
└── ...
```
## 训练
先将数据处理为lmdb格式：
```
python prepare_data.py --out LMDB ./data/ffhq/images
```
### 单机单卡
```
HIP_VISIBLE_DEVICES=2 python train.py LMDB --batch 4
```
### 单机多卡
```
HIP_VISIBLE_DEVICES=2,3 python -m torch.distributed.launch --nproc_per_node=2 --master_port=29500 train.py --batch 4 LMDB
```
## 推理
下载权重文件[stylegan2-ffhq-config-f.pt](https://pan.baidu.com/s/1IOD4DjjjrMZBF_TfL7pGJQ?pwd=1234)
SCNet快速下载链接：[stylegan2-ffhq-config-f.pt](http://113.200.138.88:18080/aimodels/findsource-dependency/stylegan2_pytorch)
模型推理：
```
python generate.py --sample 6 --pics 6 --ckpt stylegan2-ffhq-config-f.pt
```
推理结果：
<div align=center>
<img src="./assets/000000.png"/>
</div>
## 应用场景
### 算法类别
人脸生成
### 热点应用行业
安防,交通,教育
## 源码仓库及问题反馈
[https://developer.hpccube.com/codes/modelzoo/stylegan2_pytorch](https://developer.hpccube.com/codes/modelzoo/stylegan2_pytorch)
## 参考资料
[https://github.com/rosinality/stylegan2-pytorch](https://github.com/rosinality/stylegan2-pytorch)
apply_factor.py 0 → 100644
View file @ 25008cf0
import argparse
import torch
from torchvision import utils
from model import Generator
if __name__ == "__main__":
torch.set_grad_enabled(False)
parser = argparse.ArgumentParser(description="Apply closed form factorization")
parser.add_argument(
"-i", "--index", type=int, default=0, help="index of eigenvector"
)
parser.add_argument(
"-d",
"--degree",
type=float,
default=5,
help="scalar factors for moving latent vectors along eigenvector",
)
parser.add_argument(
"--channel_multiplier",
type=int,
default=2,
help='channel multiplier factor. config-f = 2, else = 1',
)
parser.add_argument("--ckpt", type=str, required=True, help="stylegan2 checkpoints")
parser.add_argument(
"--size", type=int, default=256, help="output image size of the generator"
)
parser.add_argument(
"-n", "--n_sample", type=int, default=7, help="number of samples created"
)
parser.add_argument(
"--truncation", type=float, default=0.7, help="truncation factor"
)
parser.add_argument(
"--device", type=str, default="cuda", help="device to run the model"
)
parser.add_argument(
"--out_prefix",
type=str,
default="factor",
help="filename prefix to result samples",
)
parser.add_argument(
"factor",
type=str,
help="name of the closed form factorization result factor file",
)
args = parser.parse_args()
eigvec = torch.load(args.factor)["eigvec"].to(args.device)
ckpt = torch.load(args.ckpt)
g = Generator(args.size, 512, 8, channel_multiplier=args.channel_multiplier).to(args.device)
g.load_state_dict(ckpt["g_ema"], strict=False)
trunc = g.mean_latent(4096)
latent = torch.randn(args.n_sample, 512, device=args.device)
latent = g.get_latent(latent)
direction = args.degree * eigvec[:, args.index].unsqueeze(0)
img, _ = g(
[latent],
truncation=args.truncation,
truncation_latent=trunc,
input_is_latent=True,
)
img1, _ = g(
[latent + direction],
truncation=args.truncation,
truncation_latent=trunc,
input_is_latent=True,
)
img2, _ = g(
[latent - direction],
truncation=args.truncation,
truncation_latent=trunc,
input_is_latent=True,
)
grid = utils.save_image(
torch.cat([img1, img, img2], 0),
f"{args.out_prefix}_index-{args.index}_degree-{args.degree}.png",
normalize=True,
value_range=(-1, 1),
nrow=args.n_sample,
)
calc_inception.py 0 → 100644
View file @ 25008cf0
import argparse
import pickle
import os
import torch
from torch import nn
from torch.nn import functional as F
from torch.utils.data import DataLoader
from torchvision import transforms
from torchvision.models import inception_v3, Inception3
import numpy as np
from tqdm import tqdm
from inception import InceptionV3
from dataset import MultiResolutionDataset
class Inception3Feature(Inception3):
def forward(self, x):
if x.shape[2] != 299 or x.shape[3] != 299:
x = F.interpolate(x, size=(299, 299), mode="bilinear", align_corners=True)
x = self.Conv2d_1a_3x3(x) # 299 x 299 x 3
x = self.Conv2d_2a_3x3(x) # 149 x 149 x 32
x = self.Conv2d_2b_3x3(x) # 147 x 147 x 32
x = F.max_pool2d(x, kernel_size=3, stride=2) # 147 x 147 x 64
x = self.Conv2d_3b_1x1(x) # 73 x 73 x 64
x = self.Conv2d_4a_3x3(x) # 73 x 73 x 80
x = F.max_pool2d(x, kernel_size=3, stride=2) # 71 x 71 x 192
x = self.Mixed_5b(x) # 35 x 35 x 192
x = self.Mixed_5c(x) # 35 x 35 x 256
x = self.Mixed_5d(x) # 35 x 35 x 288
x = self.Mixed_6a(x) # 35 x 35 x 288
x = self.Mixed_6b(x) # 17 x 17 x 768
x = self.Mixed_6c(x) # 17 x 17 x 768
x = self.Mixed_6d(x) # 17 x 17 x 768
x = self.Mixed_6e(x) # 17 x 17 x 768
x = self.Mixed_7a(x) # 17 x 17 x 768
x = self.Mixed_7b(x) # 8 x 8 x 1280
x = self.Mixed_7c(x) # 8 x 8 x 2048
x = F.avg_pool2d(x, kernel_size=8) # 8 x 8 x 2048
return x.view(x.shape[0], x.shape[1]) # 1 x 1 x 2048
def load_patched_inception_v3():
# inception = inception_v3(pretrained=True)
# inception_feat = Inception3Feature()
# inception_feat.load_state_dict(inception.state_dict())
inception_feat = InceptionV3([3], normalize_input=False)
return inception_feat
@torch.no_grad()
def extract_features(loader, inception, device):
pbar = tqdm(loader)
feature_list = []
for img in pbar:
img = img.to(device)
feature = inception(img)[0].view(img.shape[0], -1)
feature_list.append(feature.to("cpu"))
features = torch.cat(feature_list, 0)
return features
if __name__ == "__main__":
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
parser = argparse.ArgumentParser(
description="Calculate Inception v3 features for datasets"
)
parser.add_argument(
"--size",
type=int,
default=256,
help="image sizes used for embedding calculation",
)
parser.add_argument(
"--batch", default=64, type=int, help="batch size for inception networks"
)
parser.add_argument(
"--n_sample",
type=int,
default=50000,
help="number of samples used for embedding calculation",
)
parser.add_argument(
"--flip", action="store_true", help="apply random flipping to real images"
)
parser.add_argument("path", metavar="PATH", help="path to datset lmdb file")
args = parser.parse_args()
inception = load_patched_inception_v3()
inception = nn.DataParallel(inception).eval().to(device)
transform = transforms.Compose(
[
transforms.RandomHorizontalFlip(p=0.5 if args.flip else 0),
transforms.ToTensor(),
transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5]),
]
)
dset = MultiResolutionDataset(args.path, transform=transform, resolution=args.size)
loader = DataLoader(dset, batch_size=args.batch, num_workers=4)
features = extract_features(loader, inception, device).numpy()
features = features[: args.n_sample]
print(f"extracted {features.shape[0]} features")
mean = np.mean(features, 0)
cov = np.cov(features, rowvar=False)
name = os.path.splitext(os.path.basename(args.path))[0]
with open(f"inception_{name}.pkl", "wb") as f:
pickle.dump({"mean": mean, "cov": cov, "size": args.size, "path": args.path}, f)
closed_form_factorization.py 0 → 100644
View file @ 25008cf0
import argparse
import torch
if __name__ == "__main__":
parser = argparse.ArgumentParser(
description="Extract factor/eigenvectors of latent spaces using closed form factorization"
)
parser.add_argument(
"--out", type=str, default="factor.pt", help="name of the result factor file"
)
parser.add_argument("ckpt", type=str, help="name of the model checkpoint")
args = parser.parse_args()
ckpt = torch.load(args.ckpt)
modulate = {
k: v
for k, v in ckpt["g_ema"].items()
if "modulation" in k and "to_rgbs" not in k and "weight" in k
}
weight_mat = []
for k, v in modulate.items():
weight_mat.append(v)
W = torch.cat(weight_mat, 0)
eigvec = torch.svd(W).V.to("cpu")
torch.save({"ckpt": args.ckpt, "eigvec": eigvec}, args.out)
convert_weight.py 0 → 100644
View file @ 25008cf0
import argparse
import os
import sys
import pickle
import math
import torch
import numpy as np
from torchvision import utils
from model import Generator, Discriminator
def convert_modconv(vars, source_name, target_name, flip=False):
weight = vars[source_name + "/weight"].value().eval()
mod_weight = vars[source_name + "/mod_weight"].value().eval()
mod_bias = vars[source_name + "/mod_bias"].value().eval()
noise = vars[source_name + "/noise_strength"].value().eval()
bias = vars[source_name + "/bias"].value().eval()
dic = {
"conv.weight": np.expand_dims(weight.transpose((3, 2, 0, 1)), 0),
"conv.modulation.weight": mod_weight.transpose((1, 0)),
"conv.modulation.bias": mod_bias + 1,
"noise.weight": np.array([noise]),
"activate.bias": bias,
}
dic_torch = {}
for k, v in dic.items():
dic_torch[target_name + "." + k] = torch.from_numpy(v)
if flip:
dic_torch[target_name + ".conv.weight"] = torch.flip(
dic_torch[target_name + ".conv.weight"], [3, 4]
)
return dic_torch
def convert_conv(vars, source_name, target_name, bias=True, start=0):
weight = vars[source_name + "/weight"].value().eval()
dic = {"weight": weight.transpose((3, 2, 0, 1))}
if bias:
dic["bias"] = vars[source_name + "/bias"].value().eval()
dic_torch = {}
dic_torch[target_name + f".{start}.weight"] = torch.from_numpy(dic["weight"])
if bias:
dic_torch[target_name + f".{start + 1}.bias"] = torch.from_numpy(dic["bias"])
return dic_torch
def convert_torgb(vars, source_name, target_name):
weight = vars[source_name + "/weight"].value().eval()
mod_weight = vars[source_name + "/mod_weight"].value().eval()
mod_bias = vars[source_name + "/mod_bias"].value().eval()
bias = vars[source_name + "/bias"].value().eval()
dic = {
"conv.weight": np.expand_dims(weight.transpose((3, 2, 0, 1)), 0),
"conv.modulation.weight": mod_weight.transpose((1, 0)),
"conv.modulation.bias": mod_bias + 1,
"bias": bias.reshape((1, 3, 1, 1)),
}
dic_torch = {}
for k, v in dic.items():
dic_torch[target_name + "." + k] = torch.from_numpy(v)
return dic_torch
def convert_dense(vars, source_name, target_name):
weight = vars[source_name + "/weight"].value().eval()
bias = vars[source_name + "/bias"].value().eval()
dic = {"weight": weight.transpose((1, 0)), "bias": bias}
dic_torch = {}
for k, v in dic.items():
dic_torch[target_name + "." + k] = torch.from_numpy(v)
return dic_torch
def update(state_dict, new):
for k, v in new.items():
if k not in state_dict:
raise KeyError(k + " is not found")
if v.shape != state_dict[k].shape:
raise ValueError(f"Shape mismatch: {v.shape} vs {state_dict[k].shape}")
state_dict[k] = v
def discriminator_fill_statedict(statedict, vars, size):
log_size = int(math.log(size, 2))
update(statedict, convert_conv(vars, f"{size}x{size}/FromRGB", "convs.0"))
conv_i = 1
for i in range(log_size - 2, 0, -1):
reso = 4 * 2 ** i
update(
statedict,
convert_conv(vars, f"{reso}x{reso}/Conv0", f"convs.{conv_i}.conv1"),
)
update(
statedict,
convert_conv(
vars, f"{reso}x{reso}/Conv1_down", f"convs.{conv_i}.conv2", start=1
),
)
update(
statedict,
convert_conv(
vars, f"{reso}x{reso}/Skip", f"convs.{conv_i}.skip", start=1, bias=False
),
)
conv_i += 1
update(statedict, convert_conv(vars, f"4x4/Conv", "final_conv"))
update(statedict, convert_dense(vars, f"4x4/Dense0", "final_linear.0"))
update(statedict, convert_dense(vars, f"Output", "final_linear.1"))
return statedict
def fill_statedict(state_dict, vars, size, n_mlp):
log_size = int(math.log(size, 2))
for i in range(n_mlp):
update(state_dict, convert_dense(vars, f"G_mapping/Dense{i}", f"style.{i + 1}"))
update(
state_dict,
{
"input.input": torch.from_numpy(
vars["G_synthesis/4x4/Const/const"].value().eval()
)
},
)
update(state_dict, convert_torgb(vars, "G_synthesis/4x4/ToRGB", "to_rgb1"))
for i in range(log_size - 2):
reso = 4 * 2 ** (i + 1)
update(
state_dict,
convert_torgb(vars, f"G_synthesis/{reso}x{reso}/ToRGB", f"to_rgbs.{i}"),
)
update(state_dict, convert_modconv(vars, "G_synthesis/4x4/Conv", "conv1"))
conv_i = 0
for i in range(log_size - 2):
reso = 4 * 2 ** (i + 1)
update(
state_dict,
convert_modconv(
vars,
f"G_synthesis/{reso}x{reso}/Conv0_up",
f"convs.{conv_i}",
flip=True,
),
)
update(
state_dict,
convert_modconv(
vars, f"G_synthesis/{reso}x{reso}/Conv1", f"convs.{conv_i + 1}"
),
)
conv_i += 2
for i in range(0, (log_size - 2) * 2 + 1):
update(
state_dict,
{
f"noises.noise_{i}": torch.from_numpy(
vars[f"G_synthesis/noise{i}"].value().eval()
)
},
)
return state_dict
if __name__ == "__main__":
device = "cuda"
parser = argparse.ArgumentParser(
description="Tensorflow to pytorch model checkpoint converter"
)
parser.add_argument(
"--repo",
type=str,
required=True,
help="path to the offical StyleGAN2 repository with dnnlib/ folder",
)
parser.add_argument(
"--gen", action="store_true", help="convert the generator weights"
)
parser.add_argument(
"--disc", action="store_true", help="convert the discriminator weights"
)
parser.add_argument(
"--channel_multiplier",
type=int,
default=2,
help="channel multiplier factor. config-f = 2, else = 1",
)
parser.add_argument("path", metavar="PATH", help="path to the tensorflow weights")
args = parser.parse_args()
sys.path.append(args.repo)
import dnnlib
from dnnlib import tflib
tflib.init_tf()
with open(args.path, "rb") as f:
generator, discriminator, g_ema = pickle.load(f)
size = g_ema.output_shape[2]
n_mlp = 0
mapping_layers_names = g_ema.__getstate__()['components']['mapping'].list_layers()
for layer in mapping_layers_names:
if layer[0].startswith('Dense'):
n_mlp += 1
g = Generator(size, 512, n_mlp, channel_multiplier=args.channel_multiplier)
state_dict = g.state_dict()
state_dict = fill_statedict(state_dict, g_ema.vars, size, n_mlp)
g.load_state_dict(state_dict)
latent_avg = torch.from_numpy(g_ema.vars["dlatent_avg"].value().eval())
ckpt = {"g_ema": state_dict, "latent_avg": latent_avg}
if args.gen:
g_train = Generator(size, 512, n_mlp, channel_multiplier=args.channel_multiplier)
g_train_state = g_train.state_dict()
g_train_state = fill_statedict(g_train_state, generator.vars, size, n_mlp)
ckpt["g"] = g_train_state
if args.disc:
disc = Discriminator(size, channel_multiplier=args.channel_multiplier)
d_state = disc.state_dict()
d_state = discriminator_fill_statedict(d_state, discriminator.vars, size)
ckpt["d"] = d_state
name = os.path.splitext(os.path.basename(args.path))[0]
torch.save(ckpt, name + ".pt")
batch_size = {256: 16, 512: 9, 1024: 4}
n_sample = batch_size.get(size, 25)
g = g.to(device)
z = np.random.RandomState(0).randn(n_sample, 512).astype("float32")
with torch.no_grad():
img_pt, _ = g(
[torch.from_numpy(z).to(device)],
truncation=0.5,
truncation_latent=latent_avg.to(device),
randomize_noise=False,
)
Gs_kwargs = dnnlib.EasyDict()
Gs_kwargs.randomize_noise = False
img_tf = g_ema.run(z, None, **Gs_kwargs)
img_tf = torch.from_numpy(img_tf).to(device)
img_diff = ((img_pt + 1) / 2).clamp(0.0, 1.0) - ((img_tf.to(device) + 1) / 2).clamp(
0.0, 1.0
)
img_concat = torch.cat((img_tf, img_pt, img_diff), dim=0)
print(img_diff.abs().max())
utils.save_image(
img_concat, name + ".png", nrow=n_sample, normalize=True, range=(-1, 1)
)
dataset.py 0 → 100644
View file @ 25008cf0
from io import BytesIO
import lmdb
from PIL import Image
from torch.utils.data import Dataset
class MultiResolutionDataset(Dataset):
def __init__(self, path, transform, resolution=256):
self.env = lmdb.open(
path,
max_readers=32,
readonly=True,
lock=False,
readahead=False,
meminit=False,
)
if not self.env:
raise IOError('Cannot open lmdb dataset', path)
with self.env.begin(write=False) as txn:
self.length = int(txn.get('length'.encode('utf-8')).decode('utf-8'))
self.resolution = resolution
self.transform = transform
def __len__(self):
return self.length
def __getitem__(self, index):
with self.env.begin(write=False) as txn:
key = f'{self.resolution}-{str(index).zfill(5)}'.encode('utf-8')
img_bytes = txn.get(key)
buffer = BytesIO(img_bytes)
img = Image.open(buffer)
img = self.transform(img)
return img
distributed.py 0 → 100644
View file @ 25008cf0
import math
import pickle
import torch
from torch import distributed as dist
from torch.utils.data.sampler import Sampler
def get_rank():
if not dist.is_available():
return 0
if not dist.is_initialized():
return 0
return dist.get_rank()
def synchronize():
if not dist.is_available():
return
if not dist.is_initialized():
return
world_size = dist.get_world_size()
if world_size == 1:
return
dist.barrier()
def get_world_size():
if not dist.is_available():
return 1
if not dist.is_initialized():
return 1
return dist.get_world_size()
def reduce_sum(tensor):
if not dist.is_available():
return tensor
if not dist.is_initialized():
return tensor
tensor = tensor.clone()
dist.all_reduce(tensor, op=dist.ReduceOp.SUM)
return tensor
def gather_grad(params):
world_size = get_world_size()
if world_size == 1:
return
for param in params:
if param.grad is not None:
dist.all_reduce(param.grad.data, op=dist.ReduceOp.SUM)
param.grad.data.div_(world_size)
def all_gather(data):
world_size = get_world_size()
if world_size == 1:
return [data]
buffer = pickle.dumps(data)
storage = torch.ByteStorage.from_buffer(buffer)
tensor = torch.ByteTensor(storage).to('cuda')
local_size = torch.IntTensor([tensor.numel()]).to('cuda')
size_list = [torch.IntTensor([0]).to('cuda') for _ in range(world_size)]
dist.all_gather(size_list, local_size)
size_list = [int(size.item()) for size in size_list]
max_size = max(size_list)
tensor_list = []
for _ in size_list:
tensor_list.append(torch.ByteTensor(size=(max_size,)).to('cuda'))
if local_size != max_size:
padding = torch.ByteTensor(size=(max_size - local_size,)).to('cuda')
tensor = torch.cat((tensor, padding), 0)
dist.all_gather(tensor_list, tensor)
data_list = []
for size, tensor in zip(size_list, tensor_list):
buffer = tensor.cpu().numpy().tobytes()[:size]
data_list.append(pickle.loads(buffer))
return data_list
def reduce_loss_dict(loss_dict):
world_size = get_world_size()
if world_size < 2:
return loss_dict
with torch.no_grad():
keys = []
losses = []
for k in sorted(loss_dict.keys()):
keys.append(k)
losses.append(loss_dict[k])
losses = torch.stack(losses, 0)
dist.reduce(losses, dst=0)
if dist.get_rank() == 0:
losses /= world_size
reduced_losses = {k: v for k, v in zip(keys, losses)}
return reduced_losses
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment