mcanalesmayo / SparseAutoencoder

Sparse Autoencoder based on the Unsupervised Feature Learning and Deep Learning tutorial from the Stanford University

deep-learning sparse-autoencoder unsupervised-learning

SparseAutoencoder

Description

The autoencoder is trained with a dataset of 10000 images (IMAGES.mat) composed by 64 input units. It has a single layer with 25 units and an output layer with 64 units.

Author

Code developed by Marcos Canales Mayo, based on the Unsupervised Feature Learning and Deep Learning tutorial from the Stanford University

About

Sparse Autoencoder based on the Unsupervised Feature Learning and Deep Learning tutorial from the Stanford University

deep-learning sparse-autoencoder unsupervised-learning

MIT License

Languages

Language:MATLAB 89.6%Language:C 10.1%Language:C++ 0.3%Language:M 0.1%