PaddlePaddle GAN library, including lots of interesting applications like First-Order motion transfer, Wav2Lip, picture repair, image editing, photo2cartoon, image style transfer, GPEN, and so on.
License: Apache License 2.0
我在使用PaddleGAN的U-GAT-IT测试时出现KeyError: "No object named 'Transfroms' found in 'PREPROCESS' registry!"
没有dlib,无法安装dlib
docs/zh_CN/tutorials/pixel2style2pixel.md里的参数配置:
cd applications/
python -u tools/styleganv2.py \
--input_image <替换为输入的图像路径> \
--output_path <替换为生成图片存放的文件夹> \
--weight_path <替换为你的预训练模型路径> \
--model_type ffhq-inversion \
--seed 233 \
--size 1024 \
--style_dim 512 \
--n_mlp 8 \
--channel_multiplier 2 \
--cpu可实际所调用的tools/styleganv2.py里并无参数input_image?
PaddleGAN/ppgan/utils/logger.py
Line 21 in 1ce8016
I wanted to understand a bit about the licensing here.
RuntimeError: Error(s) in loading state_dict for Wav2Lip
源代码是使用的cpu，
#device = 'cpu'
device = 'cuda' if torch.cuda.is_available() else 'cpu'
print('Using {} for inference.'.format(device))
export那一步，把jpg和mp4放在“同一个”目录下，不提示no such file xx.mp4，提示报错no such file xx.jpg。不知道有遇到类似问题的么？谢谢。
E0302 16:59:14.194175 332123584 pybind.cc:1415] Cannot use GPU because you have installed CPU version PaddlePaddle.
If you want to use GPU, please try to install GPU version PaddlePaddle by: pip install paddlepaddle-gpu
If you only have CPU, please change CUDAPlace(0) to be CPUPlace().
代码已替换为CPUPlace()， 但是还是报错
Installation seems fine. But when I run from ppgan.apps import Photo2CartoonPredictor
ImportError Traceback (most recent call last)
in ()
----> 1 from ppgan.apps import Photo2CartoonPredictor
ImportError: cannot import name 'Photo2CartoonPredictor'
Dear authors,
Thanks for sharing a great project.
As I see you are able to train a model a photo2caroon model, but I would like to train a cartoon2photo model?
Based on your knowledge, does the cartoon2photo work well?
Thanks for your time.
code is like:
python applications/tools/animeganv2.py --input_image path/to/gakki.png --output_path path/to/animeGAN
errors are:
--------------------------------------
C++ Traceback (most recent call last):
--------------------------------------
0 paddle::imperative::Tracer::TraceOp(std::string const&, paddle::imperative::NameVarBaseMap const&, paddle::imperative::NameVarBaseMap const&, paddle::framework::AttributeMap, std::map<std::string, std::string, std::less<std::string >, std::allocator<std::pair<std::string const, std::string > > > const&)
1 paddle::imperative::Tracer::TraceOp(std::string const&, paddle::imperative::NameVarBaseMap const&, paddle::imperative::NameVarBaseMap const&, paddle::framework::AttributeMap, paddle::platform::Place const&, bool, std::map<std::string, std::string, std::less<std::string >, std::allocator<std::pair<std::string const, std::string > > > const&)
2 paddle::imperative::PreparedOp::Run(paddle::imperative::NameVarBaseMap const&, paddle::imperative::NameVarBaseMap const&, paddle::framework::AttributeMap const&)
3 std::_Function_handler<void (paddle::framework::ExecutionContext const&), paddle::framework::OpKernelRegistrarFunctor<paddle::platform::CPUPlace, false, 0ul, paddle::operators::GemmConvKernel<paddle::platform::CPUDeviceContext, float>, paddle::operators::GemmConvKernel<paddle::platform::CPUDeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&)
4 paddle::operators::GemmConvKernel<paddle::platform::CPUDeviceContext, float>::Compute(paddle::framework::ExecutionContext const&) const
5 paddle::operators::math::Im2ColFunctor<(paddle::operators::math::ColFormat)0, paddle::platform::CPUDeviceContext, float>::operator()(paddle::platform::CPUDeviceContext const&, paddle::framework::Tensor const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, paddle::framework::Tensor*, paddle::framework::DataLayout)
6 paddle::framework::SignalHandle(char const*, int)
7 paddle::platform::GetCurrentTraceBackString[abi:cxx11]()
----------------------
Error Message Summary:
----------------------
FatalError: `Segmentation fault` is detected by the operating system.
[TimeInfo: *** Aborted at 1615879561 (unix time) try "date -d @1615879561" if you are using GNU date ***]
[SignalInfo: *** SIGSEGV (@0x7f56573e42c0) received by PID 11689 (TID 0x7f60fec05740) from PID 1463698112 ***]
Segmentation fault (core dumped)
my environment:
ubuntu 18.04,python 3.8.5,paddlepaddle-gpu 2.0.0
有人脸融合的开源代码吗
ImportError: DLL load failed: 找不到指定的模块。没有找到模块，什么问题呢？
根据nn.SpectralNorm的文档描述，如果input（weight）是fc层的权重，则应设置为0；如果input（weight）是conv层的权重，则应设置为1。默认值：0。但是在ppgan/models/discriminators/nlayers.py 和ppgan/models/generators/deoldify.py中对卷积层施加SpectralNorm时均使用了默认值：0。
Python version: Python 2.7.12
Collecting imageio-ffmpeg
Downloading http://pypi.doubanio.com/packages/44/51/8a16c76b2a19ac2af82001985c80d3caca4c373528855cb27e12b39373fb/imageio-ffmpeg-0.3.0.tar.gz (13 kB)
ERROR: Package 'imageio-ffmpeg' requires a different Python: 2.7.12 not in '>=3.4'
Line 6 in 942bb21
Ai Studio高级版
! python3 -m pip install paddlepaddle-gpu==2.0.0b0 -i https://mirror.baidu.com/pypi/simple! git clone https://hub.fastgit.org/PaddlePaddle/PaddleGAN.gitcd PaddleGAN && python -u applications/tools/first-order-demo.py --driving_video datasets/unravel.flv --source_image datasets/ssc.jpg --relative --adapt_scaleTraceback (most recent call last):
File "applications/tools/first-order-demo.py", line 18, in <module>
from ppgan.apps.first_order_predictor import FirstOrderPredictor
File "/home/aistudio/PaddleGAN/ppgan/apps/__init__.py", line 2, in <module>
from .deepremaster_predictor import DeepRemasterPredictor
File "/home/aistudio/PaddleGAN/ppgan/apps/deepremaster_predictor.py", line 24, in <module>
from ppgan.models.generators.remaster import NetworkR, NetworkC
File "/home/aistudio/PaddleGAN/ppgan/models/__init__.py", line 20, in <module>
from .makeup_model import MakeupModel
File "/home/aistudio/PaddleGAN/ppgan/models/makeup_model.py", line 30, in <module>
from ..datasets.makeup_dataset import MakeupDataset
File "/home/aistudio/PaddleGAN/ppgan/datasets/__init__.py", line 15, in <module>
from .unpaired_dataset import UnpairedDataset
File "/home/aistudio/PaddleGAN/ppgan/datasets/unpaired_dataset.py", line 4, in <module>
from .base_dataset import BaseDataset, get_transform
File "/home/aistudio/PaddleGAN/ppgan/datasets/base_dataset.py", line 10, in <module>
from .transforms import transforms as T
File "/home/aistudio/PaddleGAN/ppgan/datasets/transforms/__init__.py", line 1, in <module>
from .transforms import PairedRandomCrop, PairedRandomHorizontalFlip
File "/home/aistudio/PaddleGAN/ppgan/datasets/transforms/transforms.py", line 23, in <module>
TRANSFORMS.register(T.Transpose)
AttributeError: module 'paddle.vision.transforms' has no attribute 'Transpose'
0、给定重要参数：
--checkpoint_path：D:/code/PaddleGAN-develop/wav2Lip_model/
--audio：D:/code/PaddleGAN-develop/audio_10s.wav
--face：D:/code/PaddleGAN-develop/pic/female_720.jpg
1、代码异常返回：
(1)wav2lip_predictor.py这里提示模型有问题：
(2) wav2lip_predictor.py在这里中断：
最后运行结果：
When I'm trying to build DAIN with PaddleGAN/applications/DAIN/pwcnet/correlation_op/make.sh, I get this:
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/include
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/libs
cc: error trying to exec 'cc1plus': execvp: No such file or directory
g++: error: correlation_op.cu.o: No such file or directory
I run this sh right after I cloned PaddleGAN, is there anything else I need to do before that?
能够提供requirements.txt吗，以便直到所使用的所有模块的版本
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1759/1759 [00:59<00:00, 29.40it/s]
/root/miniconda3/lib/python3.7/site-packages/scikit_image-0.15.0-py3.7-linux-x86_64.egg/skimage/util/dtype.py:135: UserWarning: Possible precision loss when converting from float32 to uint8
.format(dtypeobj_in, dtypeobj_out))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/paddle/work/github/PaddleGAN/ppgan/apps/first_order_predictor.py", line 142, in run
fps=fps)
File "/root/miniconda3/lib/python3.7/site-packages/imageio-2.5.0-py3.7.egg/imageio/core/functions.py", line 336, in mimwrite
writer = get_writer(uri, format, "I", **kwargs)
File "/root/miniconda3/lib/python3.7/site-packages/imageio-2.5.0-py3.7.egg/imageio/core/functions.py", line 174, in get_writer
request = Request(uri, "w" + mode, **kwargs)
File "/root/miniconda3/lib/python3.7/site-packages/imageio-2.5.0-py3.7.egg/imageio/core/request.py", line 126, in __init__
self._parse_uri(uri)
File "/root/miniconda3/lib/python3.7/site-packages/imageio-2.5.0-py3.7.egg/imageio/core/request.py", line 283, in _parse_uri
raise FileNotFoundError("The directory %r does not exist" % dn)
FileNotFoundError: The directory '/paddle/work/github/PaddleGAN/test/output' does not exist
0 paddle::framework::SignalHandle(char const*, int)
1 paddle::platform::GetCurrentTraceBackString()
FatalError: A serious error (Segmentation fault) is detected by the operating system. (at /paddle/paddle/fluid/platform/init.cc:303)
[TimeInfo: *** Aborted at 1607998503 (unix time) try "date -d @1607998503" if you are using GNU date ***]
[SignalInfo: *** SIGSEGV (@0x8a0b00) received by PID 30462 (TID 0x7f9e73213700) from PID 9046784 ***]
Segmentation fault
python tools/video-enhance.py --input /home/hust/PaddleGAN/applications/video/1.mp4 --process_order DAIN DeOldify EDVR --output /home/hust/output/
Traceback (most recent call last):
File "tools/video-enhance.py", line 18, in
from ppgan.apps import DAINPredictor
File "/home/hust/PaddleGAN/ppgan/apps/init.py", line 16, in
from .deepremaster_predictor import DeepRemasterPredictor
File "/home/hust/PaddleGAN/ppgan/apps/deepremaster_predictor.py", line 24, in
from ppgan.models.generators.remaster import NetworkR, NetworkC
File "/home/hust/PaddleGAN/ppgan/models/init.py", line 20, in
from .makeup_model import MakeupModel
File "/home/hust/PaddleGAN/ppgan/models/makeup_model.py", line 32, in
from ..datasets.makeup_dataset import MakeupDataset
File "/home/hust/PaddleGAN/ppgan/datasets/init.py", line 15, in
from .unpaired_dataset import UnpairedDataset
File "/home/hust/PaddleGAN/ppgan/datasets/unpaired_dataset.py", line 18, in
from .base_dataset import BaseDataset, get_transform
File "/home/hust/PaddleGAN/ppgan/datasets/base_dataset.py", line 24, in
from .transforms import transforms as T
File "/home/hust/PaddleGAN/ppgan/datasets/transforms/init.py", line 15, in
from .transforms import PairedRandomCrop, PairedRandomHorizontalFlip, Add, ResizeToScale
File "/home/hust/PaddleGAN/ppgan/datasets/transforms/transforms.py", line 22, in
import ppgan.datasets.transforms.functional as custom_F
AttributeError: module 'ppgan' has no attribute 'datasets'
can you help me? what is wrong?thinks
Hello,
I have read about the part of loading dataset for wav2lip, and I found this line:
PaddleGAN/ppgan/datasets/wav2lip_dataset.py
Line 100 in 4183009
Why do we use 80, and what is the relationship between fps and start_frame_index ?
代码有处hardcode，
不知可否将该model以及pretrain model（如果有的话）上传？@lijianshe02，谢谢！
项目：AnimeGANv2
CPU：i7-6700K
执行：python tools/main.py --config-file configs/animeganv2_pretrain.yaml
错误：Cannot use GPU because you have installed CPU version PaddlePaddle.
If you want to use GPU, please try to install GPU version PaddlePaddle by: pip install paddlepaddle-gpu
If you only have CPU, please change CUDAPlace(0) to be CPUPlace()
怎么切换呢？在main.py中加入import paddle paddle.set_device('cpu') 不得行
如何在生产环境部署支持并发的预测服务
First order motion model的任务是image animation，给定一张源图片，给定一个驱动视频，生成一段视频
假如这里的视频是摄像头采集的人像，想制作一个实时的表情迁移，难度如何？在轻薄本的CPU上运行的话算力要求大吗？
0
进行!python3 -m pip install --upgrade ppgan时，会出现ERROR: umap-learn 0.5.1 has requirement numba>=0.49, but you'll have numba 0.48.0 which is incompatible. ERROR: pynndescent 0.5.2 has requirement numba>=0.51.2, but you'll have numba 0.48.0 which is incompatible. 这样的错误。可以怎么解决？
我是在colab上面运行的。。
ImportError: cannot import name 'ParallelEnv'
我在使用PaddleGAN的U-GAT-IT测试时出现KeyError: "No object named 'Transfroms' found in 'PREPROCESS' registry!"
AttributeError: '_dlib_pybind11.rectangle' object has no attribute 'rect'
得到的结果是这样的
rectangles[[(26, 73) (167, 214)], [(229, 73) (370, 214)]]
应该是版本的问题
setup.py 中 requirements 拼写错误
After training a wgan with wgan_mnist.yaml I use the code below to test the network:
!python tools/main.py --config-file configs/wgan_mnist.yaml --evaluate-only --load output_dir/GANModel-2020-12-15-15-54/epoch_20_weight.pdparams
However, I cannot get any results in the new 'visual_train/' folder. How can I get the testing result properly?
>>> deoldify = DeOldifyPredictor()
W1027 06:42:46.683764 70588 device_context.cc:338] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 10.1, Runtime API Version: 9.0
W1027 06:42:46.730727 70588 device_context.cc:346] device: 0, cuDNN Version: 7.5.
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 853496/853496 [01:39<00:00, 8609.03it/s]
>>>
如果要用ppgan.apps.DeOldifyPredictor来处理大批量图片的上色问题，请问有没有可以加快处理速度的方法
示例demo
Model EDVR proccess start..
[03/25 11:23:58] ppgan INFO: Found /home/aistudio/.cache/ppgan/edvr_infer_model.tar
[03/25 11:23:58] ppgan INFO: Decompressing /home/aistudio/.cache/ppgan/edvr_infer_model.tar...
2021-03-25 11:23:58,229-WARNING: The old way to load inference model is deprecated. model path: /home/aistudio/.cache/ppgan/edvr_infer_model/EDVR_model.pdmodel, params path: /home/aistudio/.cache/ppgan/edvr_infer_model/EDVR_params.pdparams
0%| | 0/1427 [00:00<?, ?it/s]
Traceback (most recent call last):
File "tools/video-enhance.py", line 114, in <module>
frames_path, temp_video_path = predictor.run(temp_video_path)
File "/home/aistudio/PaddleGAN/PaddleGAN/ppgan/apps/edvr_predictor.py", line 173, in run
outs = self.base_forward(np.array(data_feed_in))
File "/home/aistudio/PaddleGAN/PaddleGAN/ppgan/apps/base_predictor.py", line 61, in base_forward
feed=feed_dict)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1110, in run
six.reraise(*sys.exc_info())
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/six.py", line 703, in reraise
raise value
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1108, in run
return_merged=return_merged)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1238, in _run_impl
use_program_cache=use_program_cache)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/fluid/executor.py", line 1328, in _run_program
[fetch_var_name])
ValueError: In user code:
File "/root/miniconda3/envs/py37_paddle1.8/lib/python3.7/site-packages/paddle/fluid/framework.py", line 2610, in append_op
attrs=kwargs.get("attrs", None))
File "/root/miniconda3/envs/py37_paddle1.8/lib/python3.7/site-packages/paddle/fluid/layer_helper.py", line 43, in append_op
return self.main_program.current_block().append_op(*args, **kwargs)
File "/root/miniconda3/envs/py37_paddle1.8/lib/python3.7/site-packages/paddle/fluid/layers/nn.py", line 7445, in reshape
"XShape": x_shape})
File "/workspace/PaddleGAN/applications/EDVR/models/edvr/edvr_model.py", line 361, in EDVRArch
L3_fea = fluid.layers.reshape(L3_fea, [-1, N, nf, shape[3]//4, shape[4]//4])
File "/workspace/PaddleGAN/applications/EDVR/models/edvr/edvr_model.py", line 436, in net
TSA_only = self.TSA_only)
File "/workspace/PaddleGAN/applications/EDVR/models/edvr/edvr.py", line 104, in build_model
out = videomodel.net(self.feature_input[0])
File "inference_model.py", line 84, in save_inference_model
infer_model.build_model()
File "inference_model.py", line 123, in <module>
save_inference_model(args)
InvalidArgumentError: The 'shape' attribute in ReshapeOp is invalid. The input tensor X'size must be divisible by known capacity of 'shape'. But received X's shape = [5, 128, 103, 135], X's size = 8899200, 'shape' is [-1, 5, 128, 102, 135], known capacity of 'shape' is -8812800.
[Hint: Expected output_shape[unk_dim_idx] * capacity == -in_size, but received output_shape[unk_dim_idx] * capacity:-8812800 != -in_size:-8899200.] (at /paddle/paddle/fluid/operators/reshape_op.cc:208)
[operator < reshape2 > error]
can i join your team PaddleGAN?
the following are my own repo about gan
https://github.com/habout632/GAN.pth
https://github.com/habout632/StarGAN2
https://github.com/habout632/GANHub
编译如果在aistudio上nvcc编译会报错，请问该怎么解决呢？
aistudio@jupyter-93077-1071352:~$ nvcc test_nvcc.cu -c -o relu_op.cu.o -ccbin cc -DPADDLE_WITH_CUDA -DEIGEN_USE_GPU -DPADDLE_USE_DSO -DPADDLE_WITH_MKLDNN -Xcompiler -fPIC -std=c++11 -Xcompiler -fPIC -w --expt-relaxed-constexpr -O3 -DNVCC -I ${include_dir}
cc: error trying to exec 'cc1plus': execvp: No such file or directory
ppgan/metric/metric_util.py:69:11: F821 undefined name '_convert_input_type_range'
img = _convert_input_type_range(img)
^
ppgan/metric/metric_util.py:76:15: F821 undefined name '_convert_output_type_range'
out_img = _convert_output_type_range(out_img, img_type)
^
2 F821 undefined name '_convert_input_type_range'
请问下举例中Face Swapping和Motion driving有什么区别呀？感觉两者用到的应该是类似的技术
File "tools/first-order-demo.py", line 18, in
from ppgan.apps.first_order_predictor import FirstOrderPredictor
ModuleNotFoundError: No module named 'ppgan'
Get error when use wav2lip
python tools/wav2lip.py --face ..\data\test.png --audio output\audio.m4a --outfile output\output.mp4
return with
Traceback (most recent call last):
File "tools/wav2lip.py", line 107, in <module>
predictor.run()
File "paddlegan\ppgan\apps\wav2lip_predictor.py", line 206, in run
mel_idx_multiplier = 80. / fps
ZeroDivisionError: float division by zero
This may caused by ".." in the image name, as the code to get format in file "paddlegan\ppgan\apps\wav2lip_predictor.py" need to be change.
def run(self):
print(self.args.face)
if not os.path.isfile(self.args.face):
raise ValueError(
'--face argument must be a valid path to video/image file')
elif self.args.face.split('.')[1] in ['jpg', 'png', 'jpeg']: # It's unsafe to get format like this when input filename involve "./" or "../"
full_frames = [cv2.imread(self.args.face)]
fps = self.args.fps蚂蚁呀嘿项目
windows10
python3.8.8
cuda_11.1
cudnn-11.2
GPU:750Ti
When I run the wgan training, I set the dataset as mnist, and the dataloader returns data like: [64,1,28, 28].
But when I set it as Cifar10, it returns [64, 32, 32, 3].
Is there something wrong with the paddlegan dataloader or I should always swap the dimension when using RGB images?
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.