def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.') parser.add_argument('--load', help='load model') parser.add_argument('--sample', action='store_true', help='view generated examples') parser.add_argument( '--data', help='a jpeg directory', default='/media/kaicao/Data/Data/Rolled/MSP/Image_Aligned') parser.add_argument('--load-size', help='size to load the original images', type=int) parser.add_argument('--crop-size', help='crop the original images', type=int) parser.add_argument( '--log_dir', help='directory to save checkout point', type=str, default= '/media/kaicao/Data/checkpoint/FingerprintSynthesis/tensorpack/AutoEncoder/' ) args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.', default='0') parser.add_argument('--model', help='model for minutiae extraction.', type=str, default='AEC_Model') parser.add_argument('--load', help='load model', default='/AutomatedLatentRecognition/models/Minutiae/AEC_net/minutiae_AEC_64_fcn_2/model-224000.index') parser.add_argument('--inference', action='store_true', help='extract minutiae on input images') parser.add_argument('--image_dir', help='a jpeg directory', default='/AutomatedLatentRecognition/Data/minutiae_cylinder_uint8') parser.add_argument('--sample_dir', help='a jpeg directory', default='/AutomatedLatentRecognition/pred_minutiae_cylinder_aug_texture/') parser.add_argument('--data', help='a jpeg directory', default='/AutomatedLatentRecognition/Data/minutiae_cylinder_uint8_MSPLatents_STFT/') parser.add_argument('--load-size', help='size to load the original images', type=int) parser.add_argument('--batch_size', help='batch size', type=int, default=128) parser.add_argument('--crop-size', help='crop the original images', type=int) parser.add_argument('--log_dir', help='directory to save checkout point', type=str, default='/AutomatedLatentRecognition/models/Minutiae/AEC_net/minutiae_AEC_64_fcn_2_Latent_STFT/') args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu if args.batch_size: opt.BATCH = args.batch_size return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.') parser.add_argument('--load', help='load model') parser.add_argument('--sample', action='store_true', help='view generated examples') parser.add_argument( '--data', help='a jpeg directory', default='/home/kaicao/Research/AutomatedLatentRecognition/Patches') parser.add_argument('--load-size', help='size to load the original images', type=int) parser.add_argument('--crop-size', help='crop the original images', type=int) parser.add_argument( '--log_dir', help='directory to save checkout point', type=str, default= '/home/kaicao/Research/AutomatedLatentRecognition/log_AutoEncoder/') args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.') parser.add_argument('--load', help='load model') parser.add_argument('--sample', type=int, default=0, help='the number of samples in the synthetic output.') parser.add_argument('--data', help='a npz file') parser.add_argument('--output', type=str) parser.add_argument('--exp_name', type=str, default=None) # parameters for model tuning. parser.add_argument('--batch_size', type=int, default=200) parser.add_argument('--z_dim', type=int, default=100) parser.add_argument('--max_epoch', type=int, default=100) parser.add_argument('--steps_per_epoch', type=int, default=1000) parser.add_argument('--num_gen_rnn', type=int, default=400) parser.add_argument('--num_gen_feature', type=int, default=100) parser.add_argument('--num_dis_layers', type=int, default=2) parser.add_argument('--num_dis_hidden', type=int, default=200) parser.add_argument('--noise', type=float, default=0.2) parser.add_argument('--optimizer', type=str, default='AdamOptimizer', choices=['GradientDescentOptimizer', 'AdamOptimizer', 'AdadeltaOptimizer']) parser.add_argument('--learning_rate', type=float, default=0.001) parser.add_argument('--l2norm', type=float, default=0.00001) args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.') parser.add_argument('--load', help='load model') parser.add_argument('--sample', action='store_true', help='view generated examples') parser.add_argument('--data', help='a jpeg directory') parser.add_argument('--load-size', help='size to load the original images', type=int) parser.add_argument('--crop-size', help='crop the original images', type=int) args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.') parser.add_argument('--model', help='model for minutiae extraction.', type=str, default='Cao_Model') parser.add_argument('--load', help='load model') parser.add_argument('--inference', action='store_true', help='extract minutiae on input images') parser.add_argument( '--image_dir', help='a jpeg directory', default= '/home/kaicao/Dropbox/Research/AutomatedLatentRecognition/Data/minutiae_cylinder_uint8' ) parser.add_argument( '--sample_dir', help='a jpeg directory', default= '/home/kaicao/Dropbox/Research/AutomatedLatentRecognition/Data/minutiae_cylinder_uint8' ) parser.add_argument( '--data', help='a jpeg directory', default= '/home/kaicao/Dropbox/Research/AutomatedLatentRecognition/Data/minutiae_cylinder_uint8' ) parser.add_argument('--load-size', help='size to load the original images', type=int) parser.add_argument('--batch_size', help='batch size', type=int) parser.add_argument('--crop-size', help='crop the original images', type=int) parser.add_argument( '--log_dir', help='directory to save checkout point', type=str, default= '/home/kaicao/Research/AutomatedLatentRecognition/log_AutoEncoder/Minutiae_Cao' ) args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu if args.batch_size: opt.BATCH = args.batch_size return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.', default='1') parser.add_argument('--load', help='load model') parser.add_argument('--enhance', action='store_true', help='enhance examples') parser.add_argument( '--test_data', help='a jpeg directory', default='/future/Data/Rolled/selected_rolled_prints/MI0479144T_07/') parser.add_argument( '--sample_dir', help='directory for generated examples', type=str, default= '/home/kaicao/Research/AutomatedLatentRecognition/Enhancement_test') parser.add_argument( '--data', help='a jpeg directory', default= '/media/kaicao/data2/AutomatedLatentRecognition/Data/enhancement_training/' ) #'/home/kaicao/Research/AutomatedLatentRecognition/Patches' parser.add_argument('--load-size', help='size to load the original images', type=int) parser.add_argument('--batch_size', help='batch size', type=int) parser.add_argument('--crop-size', help='crop the original images', type=int) parser.add_argument( '--log_dir', help='directory to save checkout point', type=str, default= '/media/kaicao/data2/AutomatedLatentRecognition/models/Enhancement/AEC_net/Enhancement_AEC_128_depth_4_STFT_2/' ) args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu if args.batch_size: opt.BATCH = args.batch_size return args
def get_args(): parser = argparse.ArgumentParser() parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.') parser.add_argument('--load', help='load model', default='model/I-WGAN_CAE/model-620000.index') parser.add_argument( '--sample_dir', help='directory for generated examples', type=str, default= '/media/kaicao/Data/Data/FingerprintSynthesis/tensorpack/I-WGAN_CAE_10M_JPEG/' ) parser.add_argument('--num_images', help='number of fingerprint images ', type=int, default=250) args = parser.parse_args() opt.use_argument(args) if args.gpu: os.environ['CUDA_VISIBLE_DEVICES'] = args.gpu return args