schimasuperbra / cedgan-interpolation

Demo code for the paper -- Spatial interpolation using conditional generative adversarial neural networks

CEDGAN for spatial interpolation

(Source code for paper) Spatial interpolation using conditional generative adversarial neural networks https://doi.org/10.1080/13658816.2019.1599122 https://www.researchgate.net/publication/332450640_Spatial_interpolation_using_conditional_generative_adversarial_neural_networks

Usage:

train: an example of training based on the 10x10 uniform sampling

%run cdcgan.py --npre 0 --niter 200 --nk 1 --ncp 100 --lr 0.00005 --cuda --dataset DEM --batchSize 64

test: an example of calling the pre-trained model (200 epoches of training) with 10x10 sampled images

%run generate.py --batchSize 64 --netG outfile_100_samples --dataset DEM --ncp 100 --outf outfile_generate_loss/100samples

Only a small dataset is provided in this Git source, please contact [email protected] or [email protected] for further collaboration

some optional parameters:

parser = argparse.ArgumentParser()

parser.add_argument('--batchSize', type=int, default=64, help='input batch size')

parser.add_argument('--imageSize', type=int, default=32, help='the height / width of the input image to network')

parser.add_argument('--nthread', type=int,default=1, help="number of workers/subprocess")

parser.add_argument('--ncp', type=int, default=100, help='size of the controlpoints')

parser.add_argument('--ngf', type=int, default=64)

parser.add_argument('--cuda', action='store_true', help='enables cuda')

parser.add_argument('--outf', default='outfile_generate_loss', help='folder to output images and model checkpoints')

parser.add_argument('--manualSeed', type=int, help='manual seed')

parser.add_argument('--dataset', default='DEM', help='which dataset to train on, DEM')

parser.add_argument('--netG', default='outfile', help="path to netG (to continue training)")

parser.add_argument('--logfile', default='outfile_generate_loss/100samples/errlog.txt', help="logfile to record error")

Citation

Please cite our paper if CEDGAN helps you in your own work:

Zhu D, Cheng X, Zhang F, et al. Spatial interpolation using conditional generative adversarial neural networks[J]. International Journal of Geographical Information Science, 2020, 34(4): 735-758.

@article{zhu2020spatial,

title={Spatial interpolation using conditional generative adversarial neural networks},

author={Zhu, Di and Cheng, Ximeng and Zhang, Fan and Yao, Xin and Gao, Yong and Liu, Yu},

journal={International Journal of Geographical Information Science},

volume={34},

number={4},

pages={735--758},

year={2020},

publisher={Taylor & Francis}

}