* gfpgan push

* gfpgan finish

* gfpgan add init

* gfpgan del recover

* 11111

* gfpgan change name

* gfpgan recover name

* Update GFPGAN.md
Co-authored-by: Nwangna11BD <79366697+wangna11BD@users.noreply.github.com>

* add the model files and tipc of GPEN

* add the model files and tipc of GPEN

* Create singan.md

* Create gradient_penalty.py

* add GradientPenalty

* add images for singan document

* add SinGANModel

* add SinGANGenerator

* add SinGANDiscriminator

* Create discriminator_singan.py

* Create generator_singan.py

* Create singan_model.py

* Create empty_dataset.py

* add EmptyDataset

* Create singan_predictor.py

* add SinGANPredictor

* Create singan.py

* create configs for singan

* add inference for singan

* create tipc config for singan

* Create python_singan_results_fp32.txt

* add tipc prepare for singan

* Update test_train_inference_python.md

* Update readme.md

* Update singan.md

* Update singan_model.py

* Update prepare.sh

* Update train_infer_python.txt

* Update prepare.sh

* Revert "add images for singan document"

This reverts commit f45fe5e55a2588611d951ae84d776c90693788df.

* Update singan.md

* update path format in configs

* update year of copyright

* modify the order of import

* Update singan_predictor.py

* update configs for singan

* Update singan_model.py

* modify urls for singan in prepare.sh

* add pretrained weight for singan

* Update singan_model.py

* Update singan.md

* Create English tutorial for SinGAN

* add photopen 

* update photopen.md

* fix photopen_model.py

* fix the ci problem
Co-authored-by: Nqingqing01 <dangqingqing@baidu.com>

* add copyright

* update comment

* fix  copyright

* fix  copyright 2

* fix  copyright 3

* fix benchmark

* add fomm and stylegan benchmark

* fix version of imageio

* add copyright of psgan, test=document_fix

* add copyright, test=document_fix

* fix benchmark

* add fomm and stylegan benchmark

* fix version of imageio

* add copyright of psgan, test=document_fix

* add LapStyle Model

* add training od first order motion
* Modify codes according to reviews
* modify single person and multi person process of fom model. 
* Git rid of sklearn, skimage

* add configuration options
e.g. python tools/main.py --c configs/stylegan_v2_256_ffhq.yaml -o total_iters=1 log_config.visiual_interval=1

* add stargan-v2 test and train code

* code normalization

* modify FAN code

* add munch in requirements.txt

* add stylegan model

* add wav2lip trainning code

* add stylegan2 and pixel2style2pixel models, predictors and documents

* add esrgan model, refine codes

* add some custom gan models

* add animeganv2 network and dataset
* animegan:refine code,add License
Co-authored-by: Nqingqing01 <dangqingqing@baidu.com>

* fix auto downloading problem of dlib weight and add s3fd face detector

* add ugatit model

* docs: update install instruction

1. paddlepaddle2.0.0rc instruction can be found from main page
2. only two steps, install paddlepaddle and paddlegan, other than 3step

* docs: update en install
* feat: add dcgan
* reform: debug with mnist datset
validate performance with different options, and original structure and setting

* refine code

* test

* fix apps

* update readme

* rm unused code

* fix apps output when input is image

* clean code

* update requirements.txt