Understand Autoencoders by implementing in TensorFlow

Reading time: 30 minutes

What are Autoencoders?

Autoencoders are a type of unsupervised neural networks. Here, the number of neurons in the output layer and the input layer is the same i.e the output and the input of this neural network is nearly the same.

Why are they used?

As we all know, neural networks contain an internal representation of the input in each of it's hidden layer. One of the hidden layers of the autoencoder, has lesser number of neurons as compared to other layers. This laye...