K.set_session(sess)
dataset = Cifar10Wrapper.load_default()
ae_folder = 'prod/cifar10_ae3_relu_6/'
encoder_weights_file = os.path.join(ae_folder, 'encoder.h5')
decoder_weights_file = os.path.join(ae_folder, 'decoder.h5')
rbm_params_file = os.path.join(
ae_folder,
'ptrbm_scheme1/ptrbm_hid2000_lr0.001_pcd25/epoch_500_rbm.h5')
# encoder_weights_file = '/home/hhu/Developer/dem/prod/cifar10_ae3_relu_6/test_ae_fe_const_balance/epoch_500_encoder.h5'
# decoder_weights_file = encoder_weights_file.replace('encoder.', 'decoder.')
# rbm_params_file = encoder_weights_file.replace('encoder.', 'rbm.')
dem = DEM.load_from_param_files(dataset.x_shape, cifar10_ae.RELU_MAX,
cifar10_ae.encode, encoder_weights_file,
cifar10_ae.decode, decoder_weights_file,
rbm_params_file)
sampler_generator = gibbs_sampler.create_sampler_generator(
dem.rbm, None, 64, 10000)
output_dir = encoder_weights_file.rsplit('/', 1)[0]
dem_trainer = DEMTrainer(sess, dataset, dem, utils.vis_cifar10, output_dir)
z_sample = dem_trainer._draw_samples(sampler_generator())
z_data, distance = find_nearest_z_data(dem.encoder, dataset.train_xs,
z_sample)
dem_trainer._save_samples(z_sample, encoder_weights_file + '.z_sample.png')
dem_trainer._save_samples(z_data, encoder_weights_file + '.z_data.png')
with open(encoder_weights_file + '.zz_distance.txt', 'w') as f:
print >> f, distance
for zd, zs in zip(z_data, z_sample):
print >> f, list(zd[:20])
from dem import DEM
if __name__ == '__main__':
import argparse
p = argparse.ArgumentParser(description="Generate GSI DEM from SmellDEM")
p.add_argument("input_file", metavar="input_file", help="Input file")
p.add_argument("output_file", metavar="output_file", help="Output file")
args = p.parse_args()
print(args.input_file, args.output_file)
DEM.readDEMandWriteGSIDEM(args.input_file, args.output_file)
from dem import DEM
if __name__ == '__main__':
import argparse
p = argparse.ArgumentParser(description="Generate SmellDEM from GSI DEM")
p.add_argument("z", metavar="z", help="z")
p.add_argument("x", metavar="x", help="x")
p.add_argument("y", metavar="y", help="y")
p.add_argument("input_file", metavar="input_file", help="Input file")
p.add_argument("output_file", metavar="output_file", help="Output file")
p.add_argument('--gzip', help='gzipped output', action="store_true")
args = p.parse_args()
print(args.x, args.y, args.z, args.input_file, args.output_file, args.gzip)
DEM.generateFromGSIDem(args.x, args.y, args.z, args.input_file, args.output_file, args.gzip)
from dem import DEM
filename = 'dem/FG-GML-5640-55-00-DEM5A-20161001.xml'
if __name__ == '__main__':
d = DEM(filename)
ulat, ulong, llat, llong = d.GetArea()
print(ulat, ulong, llat, llong)
x, y = d.GetSize()
print(x, y)
depthArray = d.GetDepth()
for yy in range(y):
for xx in range(x):
print(depthArray[xx,yy])
def import_cmd(from_dir, to_dir, x, y, z):
input_file = os.path.join(from_dir, "{z}/{x}/{y}.txt".format(x=x, y=y, z=z))
output_file = os.path.join(to_dir, "{z}/{x}/{y}.smelldem".format(x=x, y=y, z=z))
DEM.generateFromGSIDem(x, y, z, input_file, output_file)
print(output_file)
default=None,
help='Where to store samples and models')
if __name__ == '__main__':
opt = parser.parse_args()
opt.manualSeed = 666999
print(opt)
np.random.seed(opt.manualSeed)
torch.manual_seed(opt.manualSeed)
torch.cuda.manual_seed(opt.manualSeed)
assert opt.experiment is not None, 'specify output dir to avoid overwriting.'
if not os.path.exists(opt.experiment):
os.makedirs(opt.experiment)
print(opt, file=open(os.path.join(opt.experiment, 'configs.txt'), 'w'))
cudnn.benchmark = True
dataset = Cifar10Wrapper.load_default(opt.batch_size)
dem = DEM(opt)
sampler = Sampler(opt)
print(dem.net_f)
print(sampler.net_g)
opt.max_steps = 25
dem.train(opt, dataset, sampler)
# if opt.net_f and opt.net_g:
# dem.eval(dataset.train_xs, dataset.test_xs)
