train=True,
transform=transforms.ToTensor())
trainLoader = torch.utils.data.DataLoader(trainDataset,
batch_size=opts.batchSize,
shuffle=True)
testDataset = CELEBA(root=opts.root,
train=False,
transform=transforms.ToTensor())
testLoader = torch.utils.data.DataLoader(testDataset,
batch_size=opts.batchSize,
shuffle=False)
####### Create VAE model and classifier model #######
vae = VAE(nz=opts.nz, imSize=64, fSize=opts.fSize, sig=opts.sig)
classer = CLASSIFIER(imSize=64, fSize=opts.fSize)
print vae
print classer
if vae.useCUDA:
print 'using CUDA'
vae.cuda()
classer.cuda()
else:
print '\n *** NOT USING CUDA ***\n'
#eval or train
if opts.evalMode:
opts.loadCLASSER = True
opts.loadDELTAZ = True
opts.loadVAE = True
trainLoader = torch.utils.data.DataLoader(trainDataset,
batch_size=opts.batchSize,
shuffle=True)
testDataset = CELEBA(root=opts.root,
train=False,
transform=transforms.ToTensor(),
label=opts.label)
testLoader = torch.utils.data.DataLoader(testDataset,
batch_size=opts.batchSize,
shuffle=False)
print 'Data loaders ready.'
####### Create model #######
cvae = CVAE1(nz=opts.nz, imSize=64, fSize=opts.fSize)
dis = DISCRIMINATOR(imSize=64, fSize=opts.fSize)
aux = AUX(nz=opts.nz)
classer = CLASSIFIER(imSize=64, fSize=64) #for eval only!
if cvae.useCUDA:
print 'using CUDA'
cvae.cuda()
dis.cuda()
aux.cuda()
classer.cuda()
else:
print '\n *** NOT USING CUDA ***\n'
#load model is applicable
if opts.load_VAE_from is not None:
cvae.load_params(opts.load_VAE_from)
transform=transforms.ToTensor())
trainLoader = torch.utils.data.DataLoader(trainDataset,
batch_size=opts.batchSize,
shuffle=True)
testDataset = CELEBA(root=opts.root,
train=False,
transform=transforms.ToTensor())
testLoader = torch.utils.data.DataLoader(testDataset,
batch_size=opts.batchSize,
shuffle=False)
print 'Data loaders ready.'
####### Create model #######
cvae = CVAE(nz=opts.nz, imSize=64, fSize=opts.fSize)
dis = DISCRIMINATOR(imSize=64, fSize=opts.fSize)
if cvae.useCUDA:
print 'using CUDA'
cvae.cuda()
dis.cuda()
else:
print '\n *** NOT USING CUDA ***\n'
print cvae
print dis
####### Define optimizer #######
optimizerCVAE = optim.RMSprop(
cvae.parameters(),
lr=opts.lr) #specify the params that are being upated
opts = get_args() ####### Data set ####### print 'Prepare data loaders...' transform = transforms.Compose([transforms.ToTensor(), transforms.RandomHorizontalFlip()]) trainDataset = CELEBA(root=opts.root, train=True, transform=transforms.ToTensor(), label=opts.label) trainLoader = torch.utils.data.DataLoader(trainDataset, batch_size=opts.batchSize, shuffle=True) testDataset = CELEBA(root=opts.root, train=False, transform=transforms.ToTensor(), label=opts.label) testLoader = torch.utils.data.DataLoader(testDataset, batch_size=opts.batchSize, shuffle=False) print 'Data loaders ready.' ####### Create model ####### cvae = CVAE1(nz=opts.nz, imSize=64, fSize=opts.fSize) dis = DISCRIMINATOR(imSize=64, fSize=opts.fSize) aux = AUX(nz=opts.nz) classer = CLASSIFIER(imSize=64, fSize=64) #for eval only! if cvae.useCUDA: print 'using CUDA' cvae.cuda() dis.cuda() aux.cuda() classer.cuda() else: print '\n *** NOT USING CUDA ***\n' #load model is applicable if opts.load_VAE_from is not None: cvae.load_params(opts.load_VAE_from)