This repository contains the code for Linear Decoder. Using Linear decoder we will learn features on color images from STL-10 dataset. These features will be used in CNN exercise for classifying STL-10 images. The dataset can be downloaded from http://deeplearning.stanford.edu/wiki/index.php/Exercise:Learning_color_features_with_Sparse_Autoencoders
The entire code has been written in Matlab. For more information, please refer to above link.