Coursera-Machine Learning-Andrew Ng-Programming Exercise 6

【Exercise 6 Support Vector Machines】

【代码】【第一部分】

ex6.m

数据1：线性可分数据
利用无核SVM训练

数据2：
实现高斯核 
利用高斯核训练

数据3
参数选择：C、σ
高斯核训练

%% Machine Learning Online Class
%  Exercise 6 | Support Vector Machines
%
%  Instructions
%  ------------
% 
%  This file contains code that helps you get started on the
%  exercise. You will need to complete the following functions:
%
%     gaussianKernel.m
%     dataset3Params.m
%     processEmail.m
%     emailFeatures.m
%
%  For this exercise, you will not need to change any code in this file,
%  or any other files other than those mentioned above.
%

%% Initialization
clear ; close all; clc

%% =============== Part 1: Loading and Visualizing Data ================
%  We start the exercise by first loading and visualizing the dataset. 
%  The following code will load the dataset into your environment and plot
%  the data.
%

fprintf('Loading and Visualizing Data ...\n')

% Load from ex6data1: 
% You will have X, y in your environment
load('ex6data1.mat');

% Plot training data
plotData(X, y);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ==================== Part 2: Training Linear SVM ====================
%  The following code will train a linear SVM on the dataset and plot the
%  decision boundary learned.
%

% Load from ex6data1: 
% You will have X, y in your environment
load('ex6data1.mat');

fprintf('\nTraining Linear SVM ...\n')

% You should try to change the C value below and see how the decision
% boundary varies (e.g., try C = 1000)
C = 1;
model = svmTrain(X, y, C, @linearKernel, 1e-3, 20);
visualizeBoundaryLinear(X, y, model);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =============== Part 3: Implementing Gaussian Kernel ===============
%  You will now implement the Gaussian kernel to use
%  with the SVM. You should complete the code in gaussianKernel.m
%
fprintf('\nEvaluating the Gaussian Kernel ...\n')

x1 = [1 2 1]; x2 = [0 4 -1]; sigma = 2;
sim = gaussianKernel(x1, x2, sigma);

fprintf(['Gaussian Kernel between x1 = [1; 2; 1], x2 = [0; 4; -1], sigma = %f :' ...
         '\n\t%f\n(for sigma = 2, this value should be about 0.324652)\n'], sigma, sim);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =============== Part 4: Visualizing Dataset 2 ================
%  The following code will load the next dataset into your environment and 
%  plot the data. 
%

fprintf('Loading and Visualizing Data ...\n')

% Load from ex6data2: 
% You will have X, y in your environment
load('ex6data2.mat');

% Plot training data
plotData(X, y);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ========== Part 5: Training SVM with RBF Kernel (Dataset 2) ==========
%  After you have implemented the kernel, we can now use it to train the 
%  SVM classifier.
% 
fprintf('\nTraining SVM with RBF Kernel (this may take 1 to 2 minutes) ...\n');

% Load from ex6data2: 
% You will have X, y in your environment
load('ex6data2.mat');

% SVM Parameters
C = 1; sigma = 0.1;

% We set the tolerance and max_passes lower here so that the code will run
% faster. However, in practice, you will want to run the training to
% convergence.
model= svmTrain(X, y, C, @(x1, x2) gaussianKernel(x1, x2, sigma)); 
visualizeBoundary(X, y, model);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =============== Part 6: Visualizing Dataset 3 ================
%  The following code will load the next dataset into your environment and 
%  plot the data. 
%

fprintf('Loading and Visualizing Data ...\n')

% Load from ex6data3: 
% You will have X, y in your environment
load('ex6data3.mat');

% Plot training data
plotData(X, y);

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ========== Part 7: Training SVM with RBF Kernel (Dataset 3) ==========

%  This is a different dataset that you can use to experiment with. Try
%  different values of C and sigma here.
% 

% Load from ex6data3: 
% You will have X, y in your environment
load('ex6data3.mat');

% Try different SVM Parameters here
[C, sigma] = dataset3Params(X, y, Xval, yval);

% Train the SVM
model= svmTrain(X, y, C, @(x1, x2) gaussianKernel(x1, x2, sigma));
visualizeBoundary(X, y, model);

fprintf('Program paused. Press enter to continue.\n');
pause;

linearKernel.m

【待】

function sim = linearKernel(x1, x2)
%LINEARKERNEL returns a linear kernel between x1 and x2
%   sim = linearKernel(x1, x2) returns a linear kernel between x1 and x2
%   and returns the value in sim

% Ensure that x1 and x2 are column vectors
x1 = x1(:); x2 = x2(:);

% Compute the kernel
sim = x1' * x2;  % dot product

end

gussianKernel.m

套公式

function sim = gaussianKernel(x1, x2, sigma)
%RBFKERNEL returns a radial basis function kernel between x1 and x2
%   sim = gaussianKernel(x1, x2) returns a gaussian kernel between x1 and x2
%   and returns the value in sim

% Ensure that x1 and x2 are column vectors
x1 = x1(:); x2 = x2(:);

% You need to return the following variables correctly.
sim = 0;

% ====================== YOUR CODE HERE ======================
% Instructions: Fill in this function to return the similarity between x1
%               and x2 computed using a Gaussian kernel with bandwidth
%               sigma
%
%
    sim=exp(-sum((x1-x2).^2)/2/sigma/sigma);
% =============================================================
    
end

data3Params.m

两层for循环，逐对参数计算误差（预测与实际不相符的概率）

error_val(i,j)=mean(double(predictions ~= yval));

求最小值：每计算一个比较一下/内建函数最后统一找

function [C, sigma] = dataset3Params(X, y, Xval, yval)
%DATASET3PARAMS returns your choice of C and sigma for Part 3 of the exercise
%where you select the optimal (C, sigma) learning parameters to use for SVM
%with RBF kernel
%   [C, sigma] = DATASET3PARAMS(X, y, Xval, yval) returns your choice of C and 
%   sigma. You should complete this function to return the optimal C and 
%   sigma based on a cross-validation set.
%

% You need to return the following variables correctly.
C = 1;
sigma = 0.3;

% ====================== YOUR CODE HERE ======================
% Instructions: Fill in this function to return the optimal C and sigma
%               learning parameters found using the cross validation set.
%               You can use svmPredict to predict the labels on the cross
%               validation set. For example, 
%                   predictions = svmPredict(model, Xval);
%               will return the predictions on the cross validation set.
%
%  Note: You can compute the prediction error using 
%        mean(double(predictions ~= yval))
%

    C_vec=[0.01; 0.03; 0.1; 0.3; 1; 3; 10; 30];
    sigma_vec=[0.01; 0.03; 0.1; 0.3; 1; 3; 10; 30];
    
    error_val=zeros(length(C_vec),length(sigma_vec));
    % 两种方法求最值：计算一个比较一下/内建函数最后统一找
    for i=1:length(C_vec)
       C_temp=C_vec(i);
       for j=1:length(sigma_vec)
          sigma_temp=sigma_vec(j);
          model= svmTrain(X, y, C_temp, @(x1, x2) gaussianKernel(x1, x2, sigma_temp));
          predictions = svmPredict(model, Xval);
          
          error_val(i,j)=mean(double(predictions ~= yval));
          
          %{
          error_val=mean(double(predictions ~= yval));
          if exist('error_min')
              if error_val<error_min
                  error_min=error_val;
                  C=C_temp; sigma=sigma_temp;
              end
          else
              error_min=error_val;
              C=C_temp; sigma=sigma_temp;
          end
          %}
          
       end
    end
    
    [~,ind]=min(error_val(:));
    [i,j]=ind2sub(size(error_val),ind);
    C=C_vec(i);
    sigma=sigma_vec(j);
% =========================================================================

end

svmPredict.m

【待】

function pred = svmPredict(model, X)
%SVMPREDICT returns a vector of predictions using a trained SVM model
%(svmTrain). 
%   pred = SVMPREDICT(model, X) returns a vector of predictions using a 
%   trained SVM model (svmTrain). X is a mxn matrix where there each 
%   example is a row. model is a svm model returned from svmTrain.
%   predictions pred is a m x 1 column of predictions of {0, 1} values.
%

% Check if we are getting a column vector, if so, then assume that we only
% need to do prediction for a single example
if (size(X, 2) == 1)
    % Examples should be in rows
    X = X';
end

% Dataset 
m = size(X, 1);
p = zeros(m, 1);
pred = zeros(m, 1);

if strcmp(func2str(model.kernelFunction), 'linearKernel')
    % We can use the weights and bias directly if working with the 
    % linear kernel
    p = X * model.w + model.b;
elseif strfind(func2str(model.kernelFunction), 'gaussianKernel')
    % Vectorized RBF Kernel
    % This is equivalent to computing the kernel on every pair of examples
    X1 = sum(X.^2, 2);
    X2 = sum(model.X.^2, 2)';
    K = bsxfun(@plus, X1, bsxfun(@plus, X2, - 2 * X * model.X'));
    K = model.kernelFunction(1, 0) .^ K;
    K = bsxfun(@times, model.y', K);
    K = bsxfun(@times, model.alphas', K);
    p = sum(K, 2);
else
    % Other Non-linear kernel
    for i = 1:m
        prediction = 0;
        for j = 1:size(model.X, 1)
            prediction = prediction + ...
                model.alphas(j) * model.y(j) * ...
                model.kernelFunction(X(i,:)', model.X(j,:)');
        end
        p(i) = prediction + model.b;
    end
end

% Convert predictions into 0 / 1
pred(p >= 0) =  1;
pred(p <  0) =  0;

end

visualizeBoundaryLinear.m

画线性边界

function visualizeBoundaryLinear(X, y, model)
%VISUALIZEBOUNDARYLINEAR plots a linear decision boundary learned by the
%SVM
%   VISUALIZEBOUNDARYLINEAR(X, y, model) plots a linear decision boundary 
%   learned by the SVM and overlays the data on it

w = model.w;
b = model.b;
xp = linspace(min(X(:,1)), max(X(:,1)), 100);
yp = - (w(1)*xp + b)/w(2);
plotData(X, y);
hold on;
plot(xp, yp, '-b'); 
hold off

end

visualizeBoundary.m

利用contour函数画非线性边界。

-> contour本质是3D绘图，故首先meshgrid。（实际上contour函数支持向量形式的xy）。
-> 在生成的网格上，逐点计算预测值（0/1）。

-> 利用contour函数画边界。注意：由于svmPredic函数输出的是0/1，故用0.5作为临界值。参数0.5表明只画高度为0.5这“一层”的等高线，且根据函数规定，只画“一层”时需要传递两个重复值，即[0.5 0.5]。

利用imagesc函数同时作出预测值，可以看出明显的离散0/1，决策边界（白色）恰好就是01的分界处。

MATLAB中的contour函数对网格点之间的空隙进行了插值，所以可以在原本只有01的数据中找到高度为0.5的层。

去掉[0.5 0.5]这一参数，再作出边界，放大，能够看出插值的效果。

对比ex2中logistic的边界绘制函数plotDecisionBoundary：

其作图根据是θTx（而不是g(θTx)>0.5），故contour函数临界值取0（而不是0.5）

利用imagesc函数作出θTx，可以看出连续变化而不是离散01

function visualizeBoundary(X, y, model, varargin)
%VISUALIZEBOUNDARY plots a non-linear decision boundary learned by the SVM
%   VISUALIZEBOUNDARYLINEAR(X, y, model) plots a non-linear decision 
%   boundary learned by the SVM and overlays the data on it


% Plot the training data on top of the boundary
plotData(X, y)


% Make classification predictions over a grid of values
x1plot = linspace(min(X(:,1)), max(X(:,1)), 100)';
x2plot = linspace(min(X(:,2)), max(X(:,2)), 100)';
[X1, X2] = meshgrid(x1plot, x2plot);
vals = zeros(size(X1));
for i = 1:size(X1, 2)
   this_X = [X1(:, i), X2(:, i)];
   vals(:, i) = svmPredict(model, this_X);
end


% Plot the SVM boundary
hold on
contour(X1, X2, vals, [0.5 0.5], 'b');
% contour函数的x,y既可以是向量，也可以是meshgrid之后的矩阵
%{


imagesc(x1plot, x2plot, vals)
hold on
contour(x1plot, x2plot, vals,[0.5 ,0.5] , 'w','LineWidth',2);


hold on
plotData(X, y)


%}


hold off;


end

【第二部分】

ex6_spam.m

-> 实现邮件处理有关函数：转换成索引、转换成特征x，在邮件样本上验证效果
-> 引入处理好的数据集/测试集xy，利用无核svm模型训练分类器，预测
-> 列出所训练出的模型中，对预测为“垃圾邮件”最具有效力的10个词及其权重
-> 对某封具体邮件样本进行处理、喂入模型、给出预测

%% Machine Learning Online Class
%  Exercise 6 | Spam Classification with SVMs
%
%  Instructions
%  ------------
% 
%  This file contains code that helps you get started on the
%  exercise. You will need to complete the following functions:
%
%     gaussianKernel.m
%     dataset3Params.m
%     processEmail.m
%     emailFeatures.m
%
%  For this exercise, you will not need to change any code in this file,
%  or any other files other than those mentioned above.
%

%% Initialization
clear ; close all; clc

%% ==================== Part 1: Email Preprocessing ====================
%  To use an SVM to classify emails into Spam v.s. Non-Spam, you first need
%  to convert each email into a vector of features. In this part, you will
%  implement the preprocessing steps for each email. You should
%  complete the code in processEmail.m to produce a word indices vector
%  for a given email.

fprintf('\nPreprocessing sample email (emailSample1.txt)\n');

% Extract Features
file_contents = readFile('emailSample1.txt');
word_indices  = processEmail(file_contents);

% Print Stats
fprintf('Word Indices: \n');
fprintf(' %d', word_indices);
fprintf('\n\n');

fprintf('Program paused. Press enter to continue.\n');
pause;

%% ==================== Part 2: Feature Extraction ====================
%  Now, you will convert each email into a vector of features in R^n. 
%  You should complete the code in emailFeatures.m to produce a feature
%  vector for a given email.

fprintf('\nExtracting features from sample email (emailSample1.txt)\n');

% Extract Features
file_contents = readFile('emailSample1.txt');
word_indices  = processEmail(file_contents);
features      = emailFeatures(word_indices);

% Print Stats
fprintf('Length of feature vector: %d\n', length(features));
fprintf('Number of non-zero entries: %d\n', sum(features > 0));

fprintf('Program paused. Press enter to continue.\n');
pause;

%% =========== Part 3: Train Linear SVM for Spam Classification ========
%  In this section, you will train a linear classifier to determine if an
%  email is Spam or Not-Spam.

% Load the Spam Email dataset
% You will have X, y in your environment
load('spamTrain.mat');

fprintf('\nTraining Linear SVM (Spam Classification)\n')
fprintf('(this may take 1 to 2 minutes) ...\n')

C = 0.1;
model = svmTrain(X, y, C, @linearKernel);

p = svmPredict(model, X);

fprintf('Training Accuracy: %f\n', mean(double(p == y)) * 100);

%% =================== Part 4: Test Spam Classification ================
%  After training the classifier, we can evaluate it on a test set. We have
%  included a test set in spamTest.mat

% Load the test dataset
% You will have Xtest, ytest in your environment
load('spamTest.mat');

fprintf('\nEvaluating the trained Linear SVM on a test set ...\n')

p = svmPredict(model, Xtest);

fprintf('Test Accuracy: %f\n', mean(double(p == ytest)) * 100);
pause;


%% ================= Part 5: Top Predictors of Spam ====================
%  Since the model we are training is a linear SVM, we can inspect the
%  weights learned by the model to understand better how it is determining
%  whether an email is spam or not. The following code finds the words with
%  the highest weights in the classifier. Informally, the classifier
%  'thinks' that these words are the most likely indicators of spam.
%

% Sort the weights and obtin the vocabulary list
[weight, idx] = sort(model.w, 'descend');
vocabList = getVocabList();

fprintf('\nTop predictors of spam: \n');
for i = 1:15
    fprintf(' %-15s (%f) \n', vocabList{idx(i)}, weight(i));
end

fprintf('\n\n');
fprintf('\nProgram paused. Press enter to continue.\n');
pause;

%% =================== Part 6: Try Your Own Emails =====================
%  Now that you've trained the spam classifier, you can use it on your own
%  emails! In the starter code, we have included spamSample1.txt,
%  spamSample2.txt, emailSample1.txt and emailSample2.txt as examples. 
%  The following code reads in one of these emails and then uses your 
%  learned SVM classifier to determine whether the email is Spam or 
%  Not Spam

% Set the file to be read in (change this to spamSample2.txt,
% emailSample1.txt or emailSample2.txt to see different predictions on
% different emails types). Try your own emails as well!
filename = 'spamSample1.txt';

% Read and predict
file_contents = readFile(filename);
word_indices  = processEmail(file_contents);
x             = emailFeatures(word_indices);
p = svmPredict(model, x);

fprintf('\nProcessed %s\n\nSpam Classification: %d\n', filename, p);
fprintf('(1 indicates spam, 0 indicates not spam)\n\n');

processEmail.m

处理邮件，逐词检查：是否在单词列表里？若在，在push该词在单词列表的索引号，否则跳过。

（可以遍历单词列表/利用find函数）

假如一封邮件为 xxx aaa yyy bbb（预处理后），而xxx在单词列表里编号为(93) bbb为(20)，那么输出结果是[93,20]

function word_indices = processEmail(email_contents)
%PROCESSEMAIL preprocesses a the body of an email and
%returns a list of word_indices 
%   word_indices = PROCESSEMAIL(email_contents) preprocesses 
%   the body of an email and returns a list of indices of the 
%   words contained in the email. 
%

% Load Vocabulary
vocabList = getVocabList();

% Init return value
word_indices = [];

% ========================== Preprocess Email ===========================

% Find the Headers ( \n\n and remove )
% Uncomment the following lines if you are working with raw emails with the
% full headers

% hdrstart = strfind(email_contents, ([char(10) char(10)]));
% email_contents = email_contents(hdrstart(1):end);

% Lower case
email_contents = lower(email_contents);

% Strip all HTML
% Looks for any expression that starts with < and ends with > and replace
% and does not have any < or > in the tag it with a space
email_contents = regexprep(email_contents, '<[^<>]+>', ' ');

% Handle Numbers
% Look for one or more characters between 0-9
email_contents = regexprep(email_contents, '[0-9]+', 'number');

% Handle URLS
% Look for strings starting with http:// or https://
email_contents = regexprep(email_contents, ...
                           '(http|https)://[^\s]*', 'httpaddr');

% Handle Email Addresses
% Look for strings with @ in the middle
email_contents = regexprep(email_contents, '[^\s]+@[^\s]+', 'emailaddr');

% Handle $ sign
email_contents = regexprep(email_contents, '[$]+', 'dollar');


% ========================== Tokenize Email ===========================

% Output the email to screen as well
fprintf('\n==== Processed Email ====\n\n');

% Process file
l = 0;

while ~isempty(email_contents)

    % Tokenize and also get rid of any punctuation
    [str, email_contents] = ...
       strtok(email_contents, ...
              [' @$/#.-:&*+=[]?!(){},''">_<;%' char(10) char(13)]);
   
    % Remove any non alphanumeric characters
    str = regexprep(str, '[^a-zA-Z0-9]', '');

    % Stem the word 
    % (the porterStemmer sometimes has issues, so we use a try catch block)
    try str = porterStemmer(strtrim(str)); 
    catch str = ''; continue;
    end;

    % Skip the word if it is too short
    if length(str) < 1
       continue;
    end

    % Look up the word in the dictionary and add to word_indices if
    % found
    % ====================== YOUR CODE HERE ======================
    % Instructions: Fill in this function to add the index of str to
    %               word_indices if it is in the vocabulary. At this point
    %               of the code, you have a stemmed word from the email in
    %               the variable str. You should look up str in the
    %               vocabulary list (vocabList). If a match exists, you
    %               should add the index of the word to the word_indices
    %               vector. Concretely, if str = 'action', then you should
    %               look up the vocabulary list to find where in vocabList
    %               'action' appears. For example, if vocabList{18} =
    %               'action', then, you should add 18 to the word_indices 
    %               vector (e.g., word_indices = [word_indices ; 18]; ).
    % 
    % Note: vocabList{idx} returns a the word with index idx in the
    %       vocabulary list.
    % 
    % Note: You can use strcmp(str1, str2) to compare two strings (str1 and
    %       str2). It will return 1 only if the two strings are equivalent.
    %

    for i = 1: length(vocabList)
        if strcmp(str,vocabList{i})==1
            word_indices=[word_indices;i];
            break
        end
    end
  
    %{
    result=find(strcmp(str,vocabList)==1);
    if isempty(result)
    else
        word_indices=[word_indices;result];
    end
  %}

    % =============================================================


    % Print to screen, ensuring that the output lines are not too long
    if (l + length(str) + 1) > 78
        fprintf('\n');
        l = 0;
    end
    fprintf('%s ', str);
    l = l + length(str) + 1;

end

% Print footer
fprintf('\n\n=========================\n');

end

emailFeatures.m

转换成用于训练的特征x，规则如：

[1,3]   -> 1010

[2]      -> 0100

[2,3,4] -> 0111

...

可以利用MATLAB灵活的索引方式简便实现

function x = emailFeatures(word_indices)
%EMAILFEATURES takes in a word_indices vector and produces a feature vector
%from the word indices
%   x = EMAILFEATURES(word_indices) takes in a word_indices vector and 
%   produces a feature vector from the word indices. 

% Total number of words in the dictionary
n = 1899;

% You need to return the following variables correctly.
x = zeros(n, 1);

% ====================== YOUR CODE HERE ======================
% Instructions: Fill in this function to return a feature vector for the
%               given email (word_indices). To help make it easier to 
%               process the emails, we have have already pre-processed each
%               email and converted each word in the email into an index in
%               a fixed dictionary (of 1899 words). The variable
%               word_indices contains the list of indices of the words
%               which occur in one email.
% 
%               Concretely, if an email has the text:
%
%                  The quick brown fox jumped over the lazy dog.
%
%               Then, the word_indices vector for this text might look 
%               like:
%               
%                   60  100   33   44   10     53  60  58   5
%
%               where, we have mapped each word onto a number, for example:
%
%                   the   -- 60
%                   quick -- 100
%                   ...
%
%              (note: the above numbers are just an example and are not the
%               actual mappings).
%
%              Your task is take one such word_indices vector and construct
%              a binary feature vector that indicates whether a particular
%              word occurs in the email. That is, x(i) = 1 when word i
%              is present in the email. Concretely, if the word 'the' (say,
%              index 60) appears in the email, then x(60) = 1. The feature
%              vector should look like:
%
%              x = [ 0 0 0 0 1 0 0 0 ... 0 0 0 0 1 ... 0 0 0 1 0 ..];
%
%
    x(word_indices)=ones(length(word_indices),1);
% =========================================================================
    

end

2-25