eelxpeng / concrete-vae

ConcreteVAE implemented in Tensorflow

Preparation

Run python mnist_convert_to_records.py --directory {MNIST_DIRECTORY} to prepare MNIST data for use with mnist_input.py

Training

python train.py --data_dir={DIRECTORY} --dataset=[minst|cifar10]

See train.py for command line arguments

Experiments

CIFAR10 with 1 continuous dimension and 1 category 10 dimension

python train.py --data_dir={DIRECTORY} --dataset=cifar10 --cont_dim=1 --discrete_dim=10 --iters=180000

CIFAR10 with 10 continuous dimension and 1 category 10 dimension

python train.py --data_dir={DIRECTORY} --dataset=cifar10 --cont_dim=10 --discrete_dim=10 --iters=180000

Contributions

cifar10_input.py was modified from https://github.com/tensorflow/models/tree/master/tutorials/image/cifar10 mnist_input.py was modified from https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/how_tos/reading_data/fully_connected_reader.py