This repo has three different implementations of autoencoders using PyTorch for study purposes.

• The first one, "linear_autoencoder", is an autoencoder implemented through a simple neural network.
• The second one, "convolutional_autoencoder", is an autoencoder implemented through the use of convolutional layers.
• The third one, "denoising_autoencoder", is an autoencoder implementation that aims to remove noise from images.

For all the notebooks, I used the toy MNIST dataset.