Mercurial > hg > camir-aes2014

annotate reproduce_AES53rd/rerun_table3.m @ 0:e9a9cd732c1e tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
first hg version after svn
author wolffd
date Tue, 10 Feb 2015 15:05:51 +0000
parents
children
rev   line source
wolffd@0 1 % ---
wolffd@0 2 % This script trains similarity measures and shows the
wolffd@0 3 % results regarding RBM of table 3
wolffd@0 4 %
wolffd@0 5 % Feature Preprocessing with RBMs for Music Similarity Learning
wolffd@0 6 % Son N. Tran, Daniel Wolff, Tillman Weyde, Artur Garcez, AES53rd
wolffd@0 7 % conference
wolffd@0 8 %
wolffd@0 9 % please note that the RBM training is a probabilistic process, and
wolffd@0 10 % thus the papers' results can only be reproduced approximately with
wolffd@0 11 % large numbers of iterations of this script, and selection of RBMs according to
wolffd@0 12 % their training set performance.
wolffd@0 13 % Here, training is done on 20 random initialisations of RBM features ,
wolffd@0 14 % the test results corresponding to the RBM with the best training result are then
wolffd@0 15 % returned.
wolffd@0 16 %
wolffd@0 17 % The train and test performances are output in the console
wolffd@0 18 %
wolffd@0 19 % For convenicence, The precomputed RBM features are stored in the files
wolffd@0 20 % accompaining this script.
wolffd@0 21 % In order to compute new SVM features, delete these files.
wolffd@0 22 % ---
wolffd@0 23
wolffd@0 24 % ---
wolffd@0 25 % get svm results for RBM
wolffd@0 26 % ---
wolffd@0 27 svm_table3
wolffd@0 28 % svm_test_performance
wolffd@0 29 %fprintf('SVM Original Test Result (Wolff etal. 2012)=71.20 / 83.54\n');
wolffd@0 30
wolffd@0 31 % ---
wolffd@0 32 % get gradient results for RBM
wolffd@0 33 % ---
wolffd@0 34 gradient_table3