Source code associated with Spatio-temporal video autoencoder with differentiable memory, submitted to ICLR2016.
This is a demo version, to be trained on moving MNIST dataset.
This code extends the rnn package by providing a spatio-temporal convolutional version of LSTM cells.
To run this demo, you first need to install the extracunn package, which contains cuda code for SpatialConvolutionalNoBias layer and Huber gradient computation.
More details soon.