稀疏自编码深度学习的Matlab实现

【实例简介】
稀疏自编码深度学习的Matlab实现，sparse Auto coding，Matlab code
train, m/ 7% CS294A/CS294W Programming Assignment Starter Code Instructions %%% This file contains code that helps you get started on the programming assignment. You will need to complete the code in sampleIMAgEsm l sparseAutoencoder Cost m and computeNumericalGradientm l For the purpose of completing the assignment, you do mot need to change the code in this file curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod train.m∥ %%======== %6% STEP 0: Here we provide the relevant parameters values that will l allow your sparse autoencoder to get good filters; you do not need to 9 change the parameters below visibleSize =8*8; number of input units hiddensize 25 number of hidden units sparsity Param =0.01; desired average activation of the hidden units 7 (This was denoted by the greek alpha rho, which looks like a lower-case p curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod 4/57 train.,m∥ in the lecture notes) 1 ambda=0.0001 %o weight decay parameter beta 3 %o weight of sparsity penalty term %%==: 79 STEP 1: Implement sampleIMAGES After implementing sampleIMAGES, the display_network command should fo display a random sample of 200 patches from the dataset patches sampleIMAgES; display_network(patches(:, randi(size(patches, 2), 204, 1)), 8) %为产生一个204维的列向量,每一维的值为0~10000 curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod 5/57 train.m/v %中的随机数,说明是随机取204个 patch来显示 %o Obtain random parameters theta theta= initializeParameters ( hiddenSize, visibleSize) %%=============三三三三==================================== 97 STEP 2: Implement sparseAutoencoder Cost You can implement all of the components (squared error cost, weight decay term sparsity penalty) in the cost function at once, but it may be easier to do %o it step-by-step and run gradient checking (see STEP 3 after each step We curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod 6/57 train. m v b suggest implementing the sparseAutoencoder Cost function using the following steps (a) Implement forward propagation in your neural networl and implement the % squared error term of the cost function. Implement backpropagation to compute the derivatives. Then (using lambda=beta=( run gradient Checking % to verify that the calculations corresponding to the squared error cost term are correct curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod 7/57 train. m vl (b) Add in the weight decay term (in both the cost func and the derivative calculations), then re-run Gradient Checking to verify correctness l (c) Add in the sparsity penalty term, then re-run gradi Checking to verify correctness Feel free to change the training settings when debugging your %o code. (For example, reducing the training set size curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod 8/57 train m vl/ number of hidden units may make your code run faster and setting beta and/or lambda to zero may be helpful for debugging However, in your final submission of the visualized weights, please use parameters we b gave in Step 0 above coS七 grad] sparseAutoencoderCost(theta, visibleSize, hiddensize, lambda, sparsityParam, beta, patches) 二〓二二二二二二二〓二〓二〓二〓=二====〓= curer:YiBinYUyuyibintony@163.com,WuYiUniversity ning, MATLAB Code for Sparse Autoencod 9/57 train.m vlll 96% STeP 3: Gradient Checking Hint: If you are debugging your code, performing gradien checking on smaller models and smaller training sets (e. g, using only 10 training examples and 1-2 hidden units) may speed things up l First, lets make sure your numerical gradient computation is correct for a %o simple function. After you have implemented computeNume run the following checkNumericalGradiento curer:YiBinYUyuyibintony@163.com,WuYiUniversity Deep Learning, MATLAB Code for Sparse Autoencode 10/57 【实例截图】
【核心代码】