DeepLearnToolbox使用总结

转自:http://blog.csdn.net/lifeitengup/article/details/10219075

 

GitHub链接:DeepLearnToolbox

 

 

DeepLearnToolbox

 

A Matlab toolbox for Deep Learning.

Deep Learning is a new subfield of machine learning that focuses on learning deep hierarchical models of data. It is inspired by the human brain's apparent deep (layered, hierarchical) architecture. A good overview of the theory of Deep Learning theory is Learning Deep Architectures for AI

Directories included in the toolbox

NN/ - A library for Feedforward Backpropagation Neural Networks

CNN/ - A library for Convolutional Neural Networks

DBN/ - A library for Deep Belief Networks

SAE/ - A library for Stacked Auto-Encoders

CAE/ - A library for Convolutional Auto-Encoders

util/ - Utility functions used by the libraries

data/ - Data used by the examples

tests/ - unit tests to verify toolbox is working

For references on each library check REFS.md

Setup

  1. Download.
  2. addpath(genpath('DeepLearnToolbox'));

Windows下把文件夹加入 path 即可

 

[plain]   view plain copy
  1. %LiFeiteng   
  2.   
  3. path = pwd;  
  4. files = dir(path);  
  5.   
  6. for i = 1:length(files)  
  7.       
  8.     if files(i).isdir          
  9.         file = files(i).name;         
  10.         addpath([path '/' file])  
  11.         disp(['add ' file ' to path!'])  
  12.     end     
  13.       
  14. end  


我不打算解析代码,想从代码里面学算法是stupid的;有相应的论文,readlist,talk等可以去学习。

 

DeepLearnToolbox单隐藏层NN的优化策略:mini-Batch SGD

 

[plain]   view plain copy
  1. function [nn, L]  = nntrain(nn, train_x, train_y, opts, val_x, val_y)  
  2. %NNTRAIN trains a neural net  
  3. % [nn, L] = nnff(nn, x, y, opts) trains the neural network nn with input x and  
  4. % output y for opts.numepochs epochs, with minibatches of size  
  5. % opts.batchsize. Returns a neural network nn with updated activations,  
  6. % errors, weights and biases, (nn.a, nn.e, nn.W, nn.b) and L, the sum  
  7. % squared error for each training minibatch.  
  8.   
  9. assert(isfloat(train_x), 'train_x must be a float');  
  10. assert(nargin == 4 || nargin == 6,'number ofinput arguments must be 4 or 6')  
  11.   
  12. loss.train.e               = [];  
  13. loss.train.e_frac          = [];  
  14. loss.val.e                 = [];  
  15. loss.val.e_frac            = [];  
  16. opts.validation = 0;  
  17. if nargin == 6  
  18.     opts.validation = 1;  
  19. end  
  20.   
  21. fhandle = [];  
  22. if isfield(opts,'plot') && opts.plot == 1  
  23.     fhandle = figure();  
  24. end  
  25.   
  26. m = size(train_x, 1);  
  27.   
  28. batchsize = opts.batchsize;  
  29. numepochs = opts.numepochs;  
  30.   
  31. numbatches = m / batchsize;  
  32.   
  33. assert(rem(numbatches, 1) == 0, 'numbatches must be a integer');  
  34.   
  35. L = zeros(numepochs*numbatches,1);  
  36. n = 1;  
  37. for i = 1 : numepochs  
  38.     tic;  
  39.       
  40.     kk = randperm(m);  
  41.     for l = 1 : numbatches  
  42.         batch_x = train_x(kk((l - 1) * batchsize + 1 : l * batchsize), :);  
  43.           
  44.         %Add noise to input (for use in denoising autoencoder)  
  45.         if(nn.inputZeroMaskedFraction ~= 0)  
  46.             batch_x = batch_x.*(rand(size(batch_x))>nn.inputZeroMaskedFraction);  
  47.         end  
  48.           
  49.         batch_y = train_y(kk((l - 1) * batchsize + 1 : l * batchsize), :);  
  50.           
  51.         nn = nnff(nn, batch_x, batch_y);  
  52.         nn = nnbp(nn);  
  53.         nn = nnapplygrads(nn);  
  54.           
  55.         L(n) = nn.L;  
  56.           
  57.         n = n + 1;  
  58.     end  
  59.       
  60.     t = toc;  
  61.       
  62.     if ishandle(fhandle)  
  63.         if opts.validation == 1  
  64.             loss = nneval(nn, loss, train_x, train_y, val_x, val_y);  
  65.         else  
  66.             loss = nneval(nn, loss, train_x, train_y);  
  67.         end  
  68.         nnupdatefigures(nn, fhandle, loss, opts, i);  
  69.     end  
  70.           
  71.     disp(['epoch ' num2str(i) '/' num2str(opts.numepochs) '. Took ' num2str(t) ' seconds' '. Mean squared error on training set is ' num2str(mean(L((n-numbatches):(n-1))))]);  
  72.     nn.learningRate = nn.learningRate * nn.scaling_learningRate;  
  73. end  
  74. end  


1.不管是在 nntrain、nnbp还是nnapplygrads中我都没看到 对算法收敛性的判断,

 

而且在实测的过程中 有观察到 epoch过程中 mean-squared-error有 下降-上升-下降 的走势——微小抖动在SGD中 算是正常


多数还都是在下降(epoch我一般设为 10-40,这个值可能偏小;Hinton 06 science的文章代码记得epoch了200次,我跑了3天也没跑完)

在SAE/CNN等中 也没看到收敛性的判断。

2.CAE  没有完成

3.dropout的优化策略也可以选择

 

我测试了 SAE CNN等,多几次epoch(20-30),在MNIST上正确率在 97%+的样子。

其实cost-function 可以有不同的选择,如果使用 UFLDL的优化方式(固定的优化方法,传入cost-function的函数句柄),在更改cost-function上会更自由。

 

可以改进的地方:

1. mini-Bathch SGD算法 增加收敛性判断

2.增加 L-BFGS/CG等优化算法

3.完善CAE等

4.增加min KL-熵的 Sparse Autoencoder等

5.优化算法增加对 不同cost-function的支持

转载于:https://www.cnblogs.com/retrieval/articles/3452287.html

深度学习工具包 Deprecation notice. ----- This toolbox is outdated and no longer maintained. There are much better tools available for deep learning than this toolbox, e.g. [Theano](http://deeplearning.net/software/theano/), [torch](http://torch.ch/) or [tensorflow](http://www.tensorflow.org/) I would suggest you use one of the tools mentioned above rather than use this toolbox. Best, Rasmus. DeepLearnToolbox ================ A Matlab toolbox for Deep Learning. Deep Learning is a new subfield of machine learning that focuses on learning deep hierarchical models of data. It is inspired by the human brain's apparent deep (layered, hierarchical) architecture. A good overview of the theory of Deep Learning theory is [Learning Deep Architectures for AI](http://www.iro.umontreal.ca/~bengioy/papers/ftml_book.pdf) For a more informal introduction, see the following videos by Geoffrey Hinton and Andrew Ng. * [The Next Generation of Neural Networks](http://www.youtube.com/watch?v=AyzOUbkUf3M) (Hinton, 2007) * [Recent Developments in Deep Learning](http://www.youtube.com/watch?v=VdIURAu1-aU) (Hinton, 2010) * [Unsupervised Feature Learning and Deep Learning](http://www.youtube.com/watch?v=ZmNOAtZIgIk) (Ng, 2011) If you use this toolbox in your research please cite [Prediction as a candidate for learning deep hierarchical models of data](http://www2.imm.dtu.dk/pubdb/views/publication_details.php?id=6284) ``` @MASTERSTHESIS\{IMM2012-06284, author = "R. B. Palm", title = "Prediction as a candidate for learning deep hierarchical models of data", year = "2012", } ``` Contact: rasmusbergpalm at gmail dot com Directories included in the toolbox ----------------------------------- `NN/` - A library for Feedforward Backpropagation Neural Networks `CNN/` - A library for Convolutional Neural Networks `DBN/` - A library for Deep Belief Networks `SAE/` - A library for Stacked Auto-Encoders `CAE/` - A library for Convolutional Auto-Encoders `util/` - Utility functions used by the libraries `data/` - Data used by the examples `tests/` - unit tests to verify toolbox is working For references on each library check REFS.md Setup ----- 1. Download. 2. addpath(genpath('DeepLearnToolbox')); Example: Deep Belief Network --------------------- ```matlab function test_example_DBN load mnist_uint8; train_x = double(train_x) / 255; test_x = double(test_x) / 255; train_y = double(train_y); test_y = double(test_y); %% ex1 train a 100 hidden unit RBM and visualize its weights rand('state',0) dbn.sizes = [100]; opts.numepochs = 1; opts.batchsize = 100; opts.momentum = 0; opts.alpha = 1; dbn = dbnsetup(dbn, train_x, opts); dbn = dbntrain(dbn, train_x, opts); figure; visualize(dbn.rbm{1}.W'); % Visualize the RBM weights %% ex2 train a 100-100 hidden unit DBN and use its weights to initialize a NN rand('state',0) %train dbn dbn.sizes = [100 100]; opts.numepochs = 1; opts.batchsize = 100; opts.momentum = 0; opts.alpha = 1; dbn = dbnsetup(dbn, train_x, opts); dbn = dbntrain(dbn, train_x, opts); %unfold dbn to nn nn = dbnunfoldtonn(dbn, 10); nn.activation_function = 'sigm'; %train nn opts.numepochs = 1; opts.batchsize = 100; nn = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.10, 'Too big error'); ``` Example: Stacked Auto-Encoders --------------------- ```matlab function test_example_SAE load mnist_uint8; train_x = double(train_x)/255; test_x = double(test_x)/255; train_y = double(train_y); test_y = double(test_y); %% ex1 train a 100 hidden unit SDAE and use it to initialize a FFNN % Setup and train a stacked denoising autoencoder (SDAE) rand('state',0) sae = saesetup([784 100]); sae.ae{1}.activation_function = 'sigm'; sae.ae{1}.learningRate = 1; sae.ae{1}.inputZeroMaskedFraction = 0.5; opts.numepochs = 1; opts.batchsize = 100; sae = saetrain(sae, train_x, opts); visualize(sae.ae{1}.W{1}(:,2:end)') % Use the SDAE to initialize a FFNN nn = nnsetup([784 100 10]); nn.activation_function = 'sigm'; nn.learningRate = 1; nn.W{1} = sae.ae{1}.W{1}; % Train the FFNN opts.numepochs = 1; opts.batchsize = 100; nn = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.16, 'Too big error'); ``` Example: Convolutional Neural Nets --------------------- ```matlab function test_example_CNN load mnist_uint8; train_x = double(reshape(train_x',28,28,60000))/255; test_x = double(reshape(test_x',28,28,10000))/255; train_y = double(train_y'); test_y = double(test_y'); %% ex1 Train a 6c-2s-12c-2s Convolutional neural network %will run 1 epoch in about 200 second and get around 11% error. %With 100 epochs you'll get around 1.2% error rand('state',0) cnn.layers = { struct('type', 'i') %input layer struct('type', 'c', 'outputmaps', 6, 'kernelsize', 5) %convolution layer struct('type', 's', 'scale', 2) %sub sampling layer struct('type', 'c', 'outputmaps', 12, 'kernelsize', 5) %convolution layer struct('type', 's', 'scale', 2) %subsampling layer }; cnn = cnnsetup(cnn, train_x, train_y); opts.alpha = 1; opts.batchsize = 50; opts.numepochs = 1; cnn = cnntrain(cnn, train_x, train_y, opts); [er, bad] = cnntest(cnn, test_x, test_y); %plot mean squared error figure; plot(cnn.rL); assert(er<0.12, 'Too big error'); ``` Example: Neural Networks --------------------- ```matlab function test_example_NN load mnist_uint8; train_x = double(train_x) / 255; test_x = double(test_x) / 255; train_y = double(train_y); test_y = double(test_y); % normalize [train_x, mu, sigma] = zscore(train_x); test_x = normalize(test_x, mu, sigma); %% ex1 vanilla neural net rand('state',0) nn = nnsetup([784 100 10]); opts.numepochs = 1; % Number of full sweeps through data opts.batchsize = 100; % Take a mean gradient step over this many samples [nn, L] = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.08, 'Too big error'); %% ex2 neural net with L2 weight decay rand('state',0) nn = nnsetup([784 100 10]); nn.weightPenaltyL2 = 1e-4; % L2 weight decay opts.numepochs = 1; % Number of full sweeps through data opts.batchsize = 100; % Take a mean gradient step over this many samples nn = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.1, 'Too big error'); %% ex3 neural net with dropout rand('state',0) nn = nnsetup([784 100 10]); nn.dropoutFraction = 0.5; % Dropout fraction opts.numepochs = 1; % Number of full sweeps through data opts.batchsize = 100; % Take a mean gradient step over this many samples nn = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.1, 'Too big error'); %% ex4 neural net with sigmoid activation function rand('state',0) nn = nnsetup([784 100 10]); nn.activation_function = 'sigm'; % Sigmoid activation function nn.learningRate = 1; % Sigm require a lower learning rate opts.numepochs = 1; % Number of full sweeps through data opts.batchsize = 100; % Take a mean gradient step over this many samples nn = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.1, 'Too big error'); %% ex5 plotting functionality rand('state',0) nn = nnsetup([784 20 10]); opts.numepochs = 5; % Number of full sweeps through data nn.output = 'softmax'; % use softmax output opts.batchsize = 1000; % Take a mean gradient step over this many samples opts.plot = 1; % enable plotting nn = nntrain(nn, train_x, train_y, opts); [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.1, 'Too big error'); %% ex6 neural net with sigmoid activation and plotting of validation and training error % split training data into training and validation data vx = train_x(1:10000,:); tx = train_x(10001:end,:); vy = train_y(1:10000,:); ty = train_y(10001:end,:); rand('state',0) nn = nnsetup([784 20 10]); nn.output = 'softmax'; % use softmax output opts.numepochs = 5; % Number of full sweeps through data opts.batchsize = 1000; % Take a mean gradient step over this many samples opts.plot = 1; % enable plotting nn = nntrain(nn, tx, ty, opts, vx, vy); % nntrain takes validation set as last two arguments (optionally) [er, bad] = nntest(nn, test_x, test_y); assert(er < 0.1, 'Too big error'); ``` [![Bitdeli Badge](https://d2weczhvl823v0.cloudfront.net/rasmusbergpalm/deeplearntoolbox/trend.png)](https://bitdeli.com/free "Bitdeli Badge")
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值