seangal / dcgan_vae_pytorch

dcgan combined with vae in pytorch!

dcgan_vae_pytorch

dcgan combined with vae in pytorch!

this code is based on pytorch/examples and staturecrane/dcgan_vae_torch

The original artical can be found here

Requirements

• torch
• torchvision
• visdom
• (optional) lmdb

Usage

to start visdom:

python -m visdom.server

to start the training:

usage: main.py [-h] --dataset DATASET --dataroot DATAROOT [--workers WORKERS]
               [--batchSize BATCHSIZE] [--imageSize IMAGESIZE] [--nz NZ]
               [--ngf NGF] [--ndf NDF] [--niter NITER] [--saveInt SAVEINT] [--lr LR]
               [--beta1 BETA1] [--cuda] [--ngpu NGPU] [--netG NETG]
               [--netD NETD]

optional arguments:
  -h, --help            show this help message and exit
  --dataset DATASET     cifar10 | lsun | imagenet | folder | lfw
  --dataroot DATAROOT   path to dataset
  --workers WORKERS     number of data loading workers
  --batchSize BATCHSIZE
                        input batch size
  --imageSize IMAGESIZE
                        the height / width of the input image to network
  --nz NZ               size of the latent z vector
  --ngf NGF
  --ndf NDF
  --niter NITER         number of epochs to train for
  --saveInt SAVEINT     number of epochs between checkpoints
  --lr LR               learning rate, default=0.0002
  --beta1 BETA1         beta1 for adam. default=0.5
  --cuda                enables cuda
  --ngpu NGPU           number of GPUs to use
  --netG NETG           path to netG (to continue training)
  --netD NETD           path to netD (to continue training)

dcgan_vae_pytorch's Issues

Sampler not implemented yet

Currently, the VAE has only encoder and decoder, not like on staturecrane/dcgan_vae_torch. I will add this in the future but currently the code is working fine.

When optimizing netG with VAE loss, why not use netG.zero_grad()?

效果很一般。。在cifar10数据集上

AssertionError ?

im testing on a simple dataset with 2 classes:
python main.py --dataset=folder --dataroot=/data/vae_images/ --imageSize=256 --cuda
I get the error:
Traceback (most recent call last):

  File "main.py", line 291, in <module>
    errD_real = criterion(output, label)
  File "/home/jtoy/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 206, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/jtoy/anaconda3/lib/python3.6/site-packages/torch/nn/modules/loss.py", line 36, in forward
    return backend_fn(self.size_average, weight=self.weight)(input, target)
  File "/home/jtoy/anaconda3/lib/python3.6/site-packages/torch/nn/_functions/thnn/loss.py", line 22, in forward
    assert input.nelement() == target.nelement()

Any idea how to fix this or why it is happening?

assertion error: NameError: name 'dataset' is not defined

whole traceback is :
(cgan) D:\dcgan_vae_pytorch-master>python main.py --dataset DRIVE --dataroot ./data/DRIVE
Setting up a new session...
Namespace(batchSize=64, beta1=0.5, cuda=False, dataroot='./data/DRIVE', dataset='DRIVE', imageSize=64, lr=0.0002, manualSeed=None, ndf=64, netD='', netG='', ngf=64, ngpu=1, niter=25, nz=100, outf='.', saveInt=25, workers=2)
Random Seed: 3658
WARNING: You have a CUDA device, so you should probably run with --cuda
Traceback (most recent call last):
File "main.py", line 89, in
assert dataset
NameError: name 'dataset' is not defined
please help. thanks