supervised
anomaly detection, meaning you had the anomalies and the normal data labeled. You
won’t always have this luxury, and you shouldn’t expect it either because of the massive
volumes of data that can be involved. For your next example, you will be implementing
the encoder-decoder based temporal convolutional network (ED-TCN), but it will also
be an instance of supervised anomaly detection so that it can be compared to the dilated
TCN model given a similar task. However, keep in mind that since it is based on an
autoencoder framework, the ED-TCN should also be able to perform
Do'stlaringiz bilan baham: |