After data preprocessing, spatial and temporal features are extracted in CNN and recurrent autoencoder for the classification in fully connected networks. Empiric results show that the proposed model ...

Autoencoder is a neural network model which extracts the best representing features of data by performing unsupervised prelearning. Existing autoencoder approaches have been developed for applications ...

jaehyeongAN / LSTM-Autoencoder-for-Anomaly-Detection Star 5 Code Issues Pull requests python time-series tensorflow prediction lstm anomaly-detection lstm-autoencoder Updated on Mar 9, 2020 Jupyter ...