LSTM-autoencoder
LSTM-autoencoder copied to clipboard
What do you do to implement the idea of enforcing the model to memory information of more input frames by reconstructing while predicting?