Chainer Conditional DCGAN

Chainer implementation of Conditional DCGAN.

Installation

$ cp filelist.txt{.example,}
$ pip install -r requirements.txt

Usage

Training

$ python train.py -i <input filelist>

Inference

$ python generate.py -l <label> -m <generator model>

Using trained model

$ python generate.py -l <label> -m model/mnist-gen.npz

mnist-gen.npz is a model that learned MNIST data with batchsize=100, epoch=20.

References

• Conditional GAN
• DCGAN