1D CAE for reconstructing IMS Bearing signals
This convolutional encoder decoder is adapted from the project Image segmentation.
Purpose: After training on normal signals, reconstruction errors are used as HI. Maybe also use the weights to pretrain CNN