Awesome

LSGAN

Code for Least Squares Generative Adversarial Networks and On the Effectiveness of Least Squares Generative Adversarial Networks.

Related project: improved_LSGAN

Prerequisites

• Tensorflow
• SciPy

Usage

1.Download datasets

• LSUN
• HWDB1.0 (Handwritten Chinese characters)

2.Convert data

  $ python convert_lsun.py --source_dir $SOURCE_PATH --target_dir $TARGET_PATH
  $ python convert_icdar.py --source_dir $SOURCE_PATH --target_dir $TARGET_PATH

3.Train model

  $ python train_lsun.py --data_path $DATA_PATH
  $ python train_chn.py --data_dir $DATA_DIR

Citation

If you use this work in your research, please cite:

@article{arxiv1611.04076,
  author = {Xudong Mao and Qing Li and Haoran Xie and Raymond Y.K. Lau and Zhen Wang and Stephen Paul Smolley},
  title = {Least Squares Generative Adversarial Networks},
  journal = {arXiv preprint arXiv:1611.04076},
  year = {2016}
}

Results

Church

Bedroom

Kitchen

Conference

Dining