Matlab code for Non-negative Matrix Factorization combining random projections and Nesterov iterations
Matthieu PUIGT
c0ce29124c
Ajouter 'Code/ReadMe.m'
|
vor 6 Jahren | |
|---|---|---|
| Code | vor 6 Jahren | |
| Demo.m | vor 6 Jahren | |
| README.m | vor 6 Jahren | |
| README.md | vor 6 Jahren | |
| RSI_NeNMF.m | vor 6 Jahren | |
| VANILLA_NeNMF.m | vor 6 Jahren | |
| add_paths.m | vor 6 Jahren | |
| plots.m | vor 6 Jahren | |
| stop_rule.m | vor 6 Jahren |
README.m
% This is a matlab code for Non-negative Matrix Factorization via Nesterov's
% Optimal Gradient Method.
% In this project we have applied random projections to the aforementioned
% technique via two variants of random projections,namely:
% 1. Randomized Power iterations RPI NeNMF
% 2. Randomized Subspace Iterations RSI NeNMF
% HOW TO RUN THE CODE.
%
% Step 1: run the file "add_paths", to ensure all the folders and files are added to your matlab working path.
% Step 2: run the file "data_simulations", to generate your synthetic data.
% step 3: run the file "Demo" to execute the functions available in the folder "algorithms"
% Results of the execution are saved in the folder "output"
% To make plots, use the file "plots".
% Optimal Gradient Method.
% In this project we have applied random projections to the aforementioned
% technique via two variants of random projections,namely:
% 1. Randomized Power iterations RPI NeNMF
% 2. Randomized Subspace Iterations RSI NeNMF
% HOW TO RUN THE CODE.
%
% Step 1: run the file "add_paths", to ensure all the folders and files are added to your matlab working path.
% Step 2: run the file "data_simulations", to generate your synthetic data.
% step 3: run the file "Demo" to execute the functions available in the folder "algorithms"
% Results of the execution are saved in the folder "output"
% To make plots, use the file "plots".