Pytorch implementation of DCGAN for generating 32x32 images of SVHN, CIFAR10, MNIST, FashionMNIST, and USPS dataset.
Run commands (also available in scripts.sh):
| Dataset | Run command |
|---|---|
| MNIST | python main.py --dataset mnist --epochs 25 --n_channels 1 |
| Fashion MNIST | python main.py --dataset fashionmnist --epochs 50 --n_channels 1 |
| SVHN | python main.py --dataset svhn --epochs 100 --n_channels 3 |
| USPS | python main.py --dataset usps --epochs 100 --n_channels 1 |
| CIFAR10 | python main.py --dataset cifar10 --epochs 200 --n_channels 3 |
