Autoencoders are unsupervised artificial neural networks for which the input is same as the output.They work by compressing the input into a latent representation and then reconstructing the output from this representation.
This notebook shows the implementation of following autoencoders:
- Fully connected autoencoder
- Sparse autoencoder
- Convolutional autoencoder
- Denoising (convolutional) autoencoder
- Sequence-to-sequence autoencoder
- Variational autoencoder
This notebook also covers some interesting practical applications of autoencoders such as data denoising and dimensionality reduction for data visualisation.
input and output
